Coding & Development
Developer skills are the deepest part of the ecosystem and the easiest place to waste an afternoon on something broken. We run every skill here on a real codebase and note exactly where it shines and where it needs supervision.
711 skills listed · 443 fully verified · top 10 ranked →
711 skills
- 001 Tested · Works
Skill Creator Featured
The skill that builds skills — scaffold, test and optimize your own.
- 002 Tested · Works
Test-Driven Development Featured
Forces Claude into strict red-green-refactor discipline.
- 003 Tested · Works
Current (4.4+) API Platform QueryParameter filter reference that correctly steers away from soon-removed classes
- 004 Tested · Works
Finds and reconstructs real coding-agent session transcripts across Claude, Codex, OpenCode, and more.
- 005 Tested · Works
Blocks Unity muscle memory when writing s&box (Source 2 / Facepunch) C# components, Razor UI, and networking.
- 006 Tested · Works
Documents Acquia Source's non-default JSON:API path and Canvas Page API gotchas that standard Drupal knowledge gets wrong.
- 007 Tested · Works
Evidence-based build/course-correction/review advisor for software projects — cited comparables, no vibes.
- 008 Tested · Works
Vercel AI SDK v6 patterns: ToolLoopAgent, streaming, structured output, MCP integration.
- 009 Tested · Works
Reviews or builds UI with a severity-ranked checklist for semantic HTML, ARIA, stable locators, and hover/popup traps.
- 010 Tested · Works
38 rules for REST endpoint naming, errors, pagination, versioning, and API security.
- 011 Tested · Works
Guardrail checklist that turns a naive endpoint into an auth-checked, injection-safe one.
- 012 Tested · Works
Turns a project's .architecture/ folder into a structured ADR/review/gap report instead of a vague summary.
- 013 Tested · Works
Reference playbook for centralized audit logging, SIEM forwarding, and compliance log retention.
- 014 Tested · Works
Modern web security, compatibility and code-quality patterns from Lighthouse best-practices audits.
- 015 Tested · Works
Structured PR review checklist: security, performance, correctness, severity-labeled.
- 016 Works with setup
Routes agents to the right real COMSOL control path -- offline, batch, or live session.
- 017 Tested · Works
Convex backend rules: new function syntax, validators, indexed queries over filters.
- 018 Tested · Works
Scaffolds and edits Slidev decks — live code, Mermaid, exports — for tech talks and workshops.
- 019 Tested · Works
Scans your repo, interviews you, writes a lean CLAUDE.md under 80 lines.
- 020 Tested · Works
Picks a Spring Boot 4 / Java 25 architecture from a real decision matrix, not a coin flip.
- 021 Tested · Works
Wraps the @leoflores/datadog-cli npm tool with a scripted logs/metrics/trace incident-triage workflow.
- 022 Tested · Works
Walks bad identifiers through Arlo Belshee's 7-step naming process, one proposed commit at a time.
- 023 Tested · Works
Builds Frappe REST/RPC endpoints with auth, validation, and permission checks.
- 024 Tested · Works
Grounds GitHub Actions answers in live docs.github.com pages instead of stale training memory.
- 025 Tested · Works
Turns raw commits into a Conventional-Commits PR with task-doc validation and branch-flow rules, not just gh pr create.
- 026 Tested · Works
Turns hand-rolled goroutine code into leak-free errgroup pipelines with bounded, cancellable workers.
- 027 Tested · Works
Investigating Repository History
Reconstructs cited git/PR history before risky edits, with confidence-scored evidence.
- 028 Tested · Works
Mines your SpecStory coding histories into corroborated, outcome-labeled skill candidates you forge into real skills.
- 029 Tested · Works
Idempotent, HMAC-verified checkout/webhook flows for CHIP, Curlec, Xendit MY, Billplz, toyyibPay.
- 030 Tested · Works
Modern MATLAB image display, overlays, ROIs, and viewer apps via imageshow/viewer2d.
- 031 Tested · Works
Minecraft Commands & Scripting
Full execute/selector/NBT/scoreboard/RCON reference for Minecraft 1.21.x command scripting.
- 032 Tested · Works
Step-by-step contributor guide for wiring a new headless AI CLI (Aider, Qwen, etc.) into cc-multi-cli-plugin's adapter framework.
- 033 Tested · Works
Night Market Debugging Playbook
Symptom table for Claude Code hook/CI failures, each row backed by a real commit.
- 034 Tested · Works
Repo-specific npm publish playbook, incl. a fix for redacted auth URLs in agent shells.
- 035 Works with setup
React/Next.js/Angular frontend agent enforcing shadcn primitives, design tokens, and FSD-lite conventions.
- 036 Tested · Works
Slidev syntax reference for devs — layouts, code animations, exports, live-verified.
- 037 Tested · Works
Search the step.parts catalog and download checksum-verified STEP files for standard CAD parts
- 038 Tested · Works
Developer compliance oracle for Swedish bokföringslag, moms, BAS accounts, and SIE4.
- 039 Tested · Works
Stops guess-and-check fixes; enforces hypothesis-driven debugging.
- 040 Tested · Works
Prior-art search before you build: verdict is build, fork, contribute, use, or investigate.
- 041 Tested · Works
WordPress Block Theme Converter
Converts HTML/CSS/JS into a production-ready WordPress FSE block theme, then self-checks the output.
- 042 Tested · Works
Senior WP plugin engineer persona: mandatory permission_callback, nonce, and sanitize/escape rules baked in.
- 043 Tested · Works
WCAG 2.2 a11y reference that turns vague fixes into cited, spec-exact ones.
- 044 Tested · Works
Apple AdAttributionKit on iOS: impressions, postback copies, conversion values, SKAN migration
- 045 Tested · Works
Convert Python CLIs to agent-consumable --json NDJSON with a shared helper library
- 046 Tested · Works
Generates NowInAndroid-style Kotlin/Compose feature modules with Hilt and MVVM
- 047 Tested · Works
REST/GraphQL/gRPC API design playbook with OWASP API Top 10 mitigations, RFC 7807 errors, and TDD-first endpoint patterns.
- 048 Tested · Works
Applies evaluated code-review findings, escalating scope-widening or unsound fixes
- 049 Works with setup
Audits spec-to-code drift and orphaned EARS specs across a linked-intent-dev overlay.
- 050 Tested · Works
Turns an audit finding ID into a copy-paste-ready AI fix prompt with verification and rollback.
- 051 Tested · Works
Prompt-only reliability guardrails for agent coding: scope control, root-cause fixes, honest completion.
- 052 Tested · Works
Scaffolds Claude Code subagent files with ordered frontmatter and a self-critique pass
- 053 Tested · Works
Scaffolds new Claude Code SKILL.md files with correct frontmatter and structure
- 054 Tested · Works
Building Apple Platform Products
Reference-table driven xcodebuild/swift/xcrun command coverage for build, test, sign, and deploy.
- 055 Tested · Works
Runs file-based C# apps with the .NET CLI without scaffolding a full project.
- 056 Tested · Works
Pick the right Agent Relay swarm pattern and write correct workflow/YAML configs
- 057 Tested · Works
Uncle Bob's Dependency Rule as a self-contained architecture-review skill
- 058 Tested · Works
Opinionated DDD + Clean + Hexagonal architecture guide for backend services, any language.
- 059 Tested · Works
Shared vocabulary for designing deep modules: interfaces, seams, adapters, testability
- 060 Tested · Works
Reference for authoring Claude Code slash commands: frontmatter, arguments, bash and file refs
- 061 Tested · Works
Actionable LCP/INP/CLS fixes with modern APIs (fetchpriority, scheduler.yield, Speculation Rules)
- 062 Tested · Works
CQRS architecture guide with a 475-line reference file of runnable command/query/projector code templates.
- 063 Works with setup
Build a navigable tld architecture model of a codebase via the tld CLI
- 064 Tested · Works
Scaffold a Claude Code plugin: directory layout, plugin.json, and all component formats
- 065 Tested · Works
Malware/C2/PE reversing cheat-sheet for CTFs; the stream-cipher keystream trick actually paid off
- 066 Tested · Works
Decision gates for Deep Agents: backend, subagents, middleware, HITL, checkpointing.
- 067 Tested · Works
Python library API design: naming, signatures, deprecation, fail-loud error handling.
- 068 Tested · Works
Stages only the intended change, skips secrets and teammate edits, writes a conventional commit.
- 069 Tested · Works
Fetch-rebase-push contract for publishing a finished branch, with force-with-lease and an uncommitted-changes guardrail.
- 070 Works with setup
Gates feature work behind a real requirements doc, generated and linted by the ai-devkit CLI
- 071 Tested · Works
Turns a scanned route manifest into grouped, templated Markdown API docs for a pipeline stage.
- 072 Tested · Works
Turns a codebase into CODEBASE_ANALYSIS.md with file:line citations and Mermaid diagrams
- 073 Tested · Works
Forces a complete API contract template: auth, rate limits, retries, circuit breaker, idempotency, versioning.
- 074 Tested · Works
Generates and compiles MJML 4.x into cross-client-safe HTML email with real Outlook/Gmail fixes
- 075 Tested · Works
Enforces modern FastAPI/Pydantic conventions: Annotated params, response models, no Ellipsis
- 076 Tested · Works
Forces a codebase-exploration pass and a scoped problem definition before any code gets written.
- 077 Tested · Works
React technical-design rules: env/secrets, component architecture, data flow, build gates
- 078 Tested · Works
Frontier Plan OpenCode Executor
Reality-checks a frontier-model plan step-by-step so cheaper coding models don't thrash.
- 079 Tested · Works
Maps a repo into a structured repo.md with per-project criticality and sensitive-data tags
- 080 Tested · Works
Conventional Commits, branch naming, and merge-strategy standards, with 7 reference docs.
- 081 Tested · Works
Cobra+Viper CLI patterns: exit codes, stdout/stderr, flag-to-Viper binding, testable commands.
- 082 Tested · Works
Judgment-based Go style rules (control flow, declarations, nesting) linters can't enforce.
- 083 Tested · Works
Go DI guide with a project-size decision table for manual injection vs wire, dig/fx, or samber/do.
- 084 Tested · Works
Go module hygiene: safe upgrades, govulncheck audits, and modern tool pinning.
- 085 Tested · Works
Diagnoses Kubernetes failure modes before generating hardened manifests, Helm, and Kustomize
- 086 Tested · Works
MATLAB API guidance for volshow/viewer3d volume, mesh, and medical image display.
- 087 Tested · Works
Reviews MATLAB fixed-point (fi) code for perf, codegen efficiency, and correctness
- 088 Tested · Works
Scaffolds a self-contained LangChain Deep Agent or dcode CLI agent via guided interview.
- 089 Tested · Works
Custom Remotion scene transitions via TransitionPresentation, with 6 ready-made recipes.
- 090 Tested · Works
Turns a completed run or repeated workflow into a validated reusable skill package
- 091 Tested · Works
Turns run evidence into a plan-only, approval-gated upgrade for an existing skill
- 092 Tested · Works
Generates a structured, fact-labeled prompt to resume work in a fresh agent session
- 093 Tested · Works
House style and correctness guardrails for Anchor/Quasar Solana programs and their TS clients.
- 094 Tested · Works
Wire Sponsio runtime contracts into an LLM agent: author, validate, observe, enforce via CLI
- 095 Tested · Works
Playbooks and a decision matrix for picking/reviewing MVVM, MVI, TCA, Clean, VIPER, Reactive, MVP, or Coordinator in Swift.
- 096 Tested · Works
RESHADED/4S frameworks, capacity-estimation cheatsheet, and tradeoff tables for system-design interviews.
- 097 Tested · Works
Single-particle tracking in microscopy video: locate, link, filter, MSD diffusion with trackpy
- 098 Tested · Works
WCAG 2.2 AA review of web pages: alt text, contrast math, chart and form patterns
- 099 Tested · Works
iOS AccessorySetupKit picker, discovery descriptors, and CoreBluetooth/Wi-Fi handoff
- 100 Tested · Works
iOS Live Activity and Dynamic Island patterns plus a 20-item ActivityKit review checklist
- 101 Tested · Works
Backend code-review criteria: anti-patterns, fail-fast errors, Rule of Three, quality gates
- 102 Tested · Works
Build AI chat UIs in Next.js with Vercel's ai-elements component registry
- 103 Tested · Works
Structured codebase-onboarding checklist that produces a consistent, evidence-gated project report
- 104 Tested · Works
Audits Spring/REST controllers for HTTP verbs, versioning, status codes, and compatibility
- 105 Tested · Works
Decode, analyze, and repackage Android APKs with apktool for security review
- 106 Tested · Works
Production-grade ArgoCD 3.x + Helm 3.x patterns: Application manifests, ApplicationSets, canary rollouts, secrets.
- 107 Tested · Works
Writes and rewrites SKILL.md and agent files to project conventions
- 108 Tested · Works
CloudFormation patterns for DynamoDB tables, GSIs, TTL, streams, and auto-scaling.
- 109 Tested · Works
CloudFormation patterns and limits for production Lambda: layers, DLQ, API Gateway.
- 110 Tested · Works
Reference-backed backend guidance: APIs, auth, DB, security, DevOps — self-contained, no CLI.
- 111 Tested · Works
Claude Code hook, skill, command and plugin.json contracts verified against a real repo
- 112 Tested · Works
Cloudflare platform reference: Workers, Pages, KV/D1/R2, Workers AI, WAF, IaC configs
- 113 Tested · Works
Structured request-review-merge workflow with an anti-bloat checklist and persistent review memory.
- 114 Tested · Works
A disciplined, five-principle process for simplifying code readability without touching behavior.
- 115 Tested · Works
Language-agnostic coding principles for maintainability, readability, and code quality
- 116 Tested · Works
Wires up the sponsio-claude-code plugin so tool calls get contract-guarded and blocked
- 117 Tested · Works
Drizzle schemas for Cloudflare D1 with correct boolean, timestamp, JSON and param limits
- 118 Tested · Works
Swift DB driver patterns: wire protocol, NIO pooling, backpressure, state machines.
- 119 Tested · Works
Multi-round, bias-audited research process for costly-to-reverse tech and architecture picks.
- 120 Works with setup
Forces a 2-4 option pros/cons/criteria matrix before any architecture or refactor code gets written.
- 121 Tested · Works
Structured design-review phase: cross-checks a design doc against its requirements before you build.
- 122 Tested · Works
Reference for hardened Electron desktop apps: IPC, tray, PTY, WebRTC, and signed builds
- 123 Works with setup
Scaffolds a repo-local autonomous coding-agent loop for the flow-next Ralph harness.
- 124 Tested · Works
Merges scan-deps, secrets, and code findings into one prioritized security report
- 125 Tested · Works
Atomic commit splitting, safe rebase/squash, and git history archaeology in one skill.
- 126 Tested · Works
Turns plain-English requests into validated Claude Code / Codex / OpenCode hook configs.
- 127 Tested · Works
KubeSphere DevOps Jenkins Configuration
Kubectl/CasC playbook for KubeSphere's embedded Jenkins: auth, agents, builds, and troubleshooting.
- 128 Tested · Works
Reviews and refactors Rails code by architecture layer using the specification test
- 129 Tested · Works
Analyzes an existing codebase and generates CLAUDE.md and Claude Code config
- 130 Tested · Works
Build MCP servers that connect Claude to your tools.
- 131 Tested · Works
Migrating To Swift Concurrency
Apple's official Swift 6 concurrency migration guide, packaged as a skill
- 132 Tested · Works
Splits implementation plans into parallel subagent tasks with mandatory review gates.
- 133 Tested · Works
Diagnose NanoClaw's containerized agent runtime via its two SQLite session DBs and logs.
- 134 Tested · Works
Night Market Architecture Contract
Architecture invariants, ADRs and known weak points — scoped to the claude-night-market repo
- 135 Tested · Works
100-skill Claude Code plugin for outline-driven, diagram-first, atomic-commit engineering
- 136 Tested · Works
Migrates Claude API model strings and prompts from Sonnet 4.x / Opus 4.1 to Opus 4.5
- 137 Tested · Works
OWASP Top 10:2025, ASVS 5.0, and LLM/agentic security guidance for code review
- 138 Works with setup
Claude orchestrates Codex CLI to implement a plan, reviews the diff, loops until it passes.
- 139 Tested · Works
Builds a repo-docs guide from one real code path, with an evidence table and a validator
- 140 Tested · Works
Builds Chinese repo docs from real code runs; English paths and symbols stay exact
- 141 Tested · Works
Steers Rust code toward idiomatic error handling, borrowing, iterators, and async patterns
- 142 Tested · Works
Keeps Claude on idiomatic Synalinks API instead of Keras/DSPy/LangChain guesswork
- 143 Tested · Works
Scaffolds a real Vite + shadcn/ui React app and bundles it into one portable HTML artifact.
- 144 Tested · Works
Turns Codex/Claude Code session traces into a private, evidence-tiered retrospective with an actionable improvement queue.
- 145 Tested · Works
Removes AI-generated code bloat in 7 test-verified passes, reverting any pass that breaks tests
- 146 Tested · Works
House REST rules Claude applies when designing endpoints: URL shape, data/error/meta, 422
- 147 Tested · Works
Command reference for asdf-vm: plugins, .tool-versions, shims, and migrating off nvm/pyenv/rbenv.
- 148 Tested · Works
Bundled AgentCore reference docs: Gateway, Runtime, Memory, Registry, Evaluations CLI patterns
- 149 Tested · Works
Configures AWS MCP servers: hosted no-auth docs endpoint or uvx proxy with credentials
- 150 Tested · Works
Curated Microsoft Learn index for Azure AD B2C, fetched live instead of recalled
- 151 Tested · Works
Routes Azure Advisor dev tasks to the exact Microsoft Learn docs and fetches them
- 152 Tested · Works
Azure AI Search/Speech/OpenAI router with condensed per-language SDK cheatsheets
- 153 Tested · Works
Routes Azure AI Vision dev questions to authoritative Microsoft Learn docs and fetches them
- 154 Tested · Works
Plan-first azd project prep: generates azure.yaml, secure Bicep/Terraform, Dockerfiles
- 155 Tested · Works
Records a read-only CDP trace of a browser run and bisects it into per-page searchable buckets
- 156 Works with setup
Scans a codebase and drafts a tight CLAUDE.md of only non-obvious commands and gotchas
- 157 Tested · Works
Reviews AI-written code against Clean Code, SOLID and 15 documented LLM failure modes
- 158 Works with setup
Language-agnostic anti-pattern, type-safety, and test-first standards for reviews and refactors
- 159 Tested · Works
Minimal R8/ProGuard setup for Compose release builds; rejects over-broad keep rules
- 160 Works with setup
Pulls step-level CircleCI job logs via the v1.1 API and prints only the failed steps
- 161 Tested · Works
Halts on a failing command and fixes the root cause instead of bypassing it
- 162 Tested · Works
Scaffolds a FastAPI + React/Vite/shadcn starter with OpenAI chat wired in.
- 163 Works with setup
Cherry Studio's in-house PR/branch reviewer enforcing DataApi/IpcApi and layer-boundary rules.
- 164 Tested · Works
Drafts Keep-a-Changelog release notes, picks the semver bump, prints the gh release command
- 165 Tested · Works
Structured gh-CLI workflow to fix a GitHub issue end-to-end and open a PR
- 166 Tested · Works
Guidance for writing Claude Code plugin hooks across all nine event types
- 167 Tested · Works
Reference for bundling MCP servers (stdio/SSE/HTTP/ws) into Claude Code plugins
- 168 Tested · Works
LoRA/QLoRA fine-tuning code for HuggingFace models with rank/alpha guidance and OOM fixes.
- 169 Tested · Works
Builds PinMe Cloudflare Worker code that accepts payments through UniwebPay.
- 170 Tested · Works
Contributor workflow for the Elixir/Phoenix plugin repo: eval, lint, test before commit
- 171 Tested · Works
Logs corrections, errors, and feature gaps into a local .learnings knowledge base
- 172 Tested · Works
Audits Swift Package graphs and pbxproj links to find what actually slows Xcode builds
- 173 Tested · Works
Keeps CLAUDE.md, GEMINI.md and AGENTS.md identical by copying the changed one to the others
- 174 Tested · Works
Turns repeated code into one abstraction: fixed pipeline, pluggable variants, Mermaid flowchart
- 175 Tested · Works
Step-by-step checklist for adding a new command to the Doist todoist-cli codebase
- 176 Works with setup
Guides adding a data source or connector to a Power Apps Canvas App, then verifies via MCP
- 177 Works with setup
Agent Rules (AGENTS.md Generator)
Detects your stack and generates root + per-directory AGENTS.md files with a bundled verification pipeline.
- 178 Works with setup
Documents existing API routes into markdown/OpenAPI; does not scaffold new endpoint code.
- 179 Works with setup
Project-wide technical-debt scan ranked by impact vs refactor effort into a report
- 180 Works with setup
Orchestrates a four-phase Jetpack Compose performance audit across 25 focused skills
- 181 Works with setup
Edit Simulink, Stateflow, Simscape and System Composer models via MATLAB MCP tools
- 182 Works with setup
Taze-powered dependency bumps: structured plan, batch major decisions, never silent.
- 183 Tested · Works
Generates and audits CLAUDE.md files against a fixed section checklist and quality score
- 184 Works with setup
Clones a reference site into Hydrogen+Weaverse sections behind a mandatory approval checkpoint.
- 185 Tested · Works
CloudFormation template patterns, stack CLI commands, drift and rollback troubleshooting
- 186 Tested · Works
Disciplined multi-file rename/replace refactors with edge-case and breaking-API guardrails
- 187 Tested · Works
Reviews Java concurrent code for races, deadlocks, visibility, and virtual-thread patterns
- 188 Tested · Works
Autonomous multi-agent pipeline: research, plan, implement, test, and open a PR for a whole ticket with zero checkpoints.
- 189 Works with setup
One Bun CLI to generate images across 8 providers (OpenAI, Gemini, Flux, Stability, more)
- 190 Works with setup
Turns Odoo upgrade logs and addon scans into a classified worklist via the odoo-mcp server
- 191 Works with setup
Turns a feature request into a TDD task-and-dependency plan for autonomous parallel execution.
- 192 Works with setup
Connect to RoadRunner from MATLAB and manage project/scene/scenario lifecycle
- 193 Tested · Works
React/Next.js/TS cheat-sheet that reliably adds a11y and structure to frontend output.
- 194 Works with setup
Routes Shipyard's parallel-agent build/audit/ship workflows to sequential personas under Codex.
- 195 Works with setup
Tears down Rails dual-boot scaffolding and NextRails branches after an upgrade
- 196 Works with setup
Build-and-run loop for WinUI 3 apps: BuildAndRun.ps1, winapp run, build-error fix table
- 197 Works with setup
asc CLI commands for finding latest builds, checking processing, and expiring old ones
- 198 Works with setup
Executes azd/terraform/az deployments for already-prepared Azure apps with error recovery.
- 199 Works with setup
Better Auth setup for TypeScript apps: email/password, OAuth, 2FA, passkeys, DB adapters
- 200 Works with setup
Snapshots R/Python/Stata replication environments into lockfiles, seeds, and a paste-ready requirements block.
- 201 Tested · Works
Read-only code review with a 4-part checklist and a severity-graded report format
- 202 Works with setup
Search local coding-agent session history before acting, with citation discipline
- 203 Works with setup
Slash command that walks fixing a GitHub issue with branch, approval gates, and PR handoff
- 204 Works with setup
Orchestrator spawning isolated code and eval subagents to build spec manifests via TDD
- 205 Works with setup
Runs an implementation plan phase by phase, delegating each phase to specialist agents
- 206 Works with setup
Drive Compose HotSwan from an AI agent: edit, hot-reload, screenshot, iterate via MCP
- 207 Works with setup
Hands coding tasks to Google's Jules agent via its CLI or REST API; needs a Google login
- 208 Works with setup
Debug LangChain/LangGraph agents by fetching LangSmith traces via the langsmith-fetch CLI
- 209 Tested · Works
Bootstraps a claude-night-market dev environment: uv-first setup, Python tiers, known traps
- 210 Works with setup
Offensive TOCTOU (defensive testing)
Exploits time-of-check/time-of-use race conditions across binaries, kernels, and web apps.
- 211 Works with setup
Backend agent: router/service/repository layering, explicit transactions, ORM N+1 rules
- 212 Works with setup
Ingest a geo file into a serverless PortalJS portal as PMTiles + GeoParquet tiers
- 213 Works with setup
Fans out parallel sub-agents to review a branch/PR in the win-dev-skills repo
- 214 Works with setup
Dispatches parallel AI workers into isolated git worktrees via the subtask CLI
- 215 Works with setup
Measures Xcode clean, cached-clean and incremental build times into a JSON artifact
- 216 Works with setup
Applies approved Xcode build-setting, script-phase and Swift fixes, then re-benchmarks
- 217 Works with setup
Wires the Azure DevOps connector into a Power Apps code app via the pa CLI
- 218 Works with setup
Designs REST or GraphQL APIs with resource models, versioning, and error standards.
- 219 Works with setup
AWS API Gateway reference: REST/HTTP CLI recipes, CORS, JWT auth, 502/504 troubleshooting
- 220 Works with setup
Aws Cloudformation Auto Scaling
CloudFormation scaffolding for EC2/ECS Auto Scaling groups, policies, and lifecycle hooks
- 221 Works with setup
Line-number code insertion via a Python helper; batch mode needs uninstalled runtime
- 222 Works with setup
Multi-agent codebase recon workflow that needs a custom Team/Task tool stack, not stock Claude Code.
- 223 Works with setup
Generates env-driven nginx/Postgres/Redis configs, but its own validation phase needs sibling skills.
- 224 Works with setup
Delegates coding tasks to xAI's grok CLI headlessly, reviews every diff before commit
- 225 Works with setup
Gated multi-phase workflow to build a new component from a GitHub issue in AgnosticUI
- 226 Works with setup
Orchestrates nWave's DELIVER wave: roadmap, TDD execution, review, finalize via sub-agents
- 227 Works with setup
Orchestrates evo's parallel autoresearch loop to hill-climb a codebase benchmark
- 228 Works with setup
Creates a draft PR, monitors GitHub Actions CI, and fixes failures until checks pass
- 229 Works with setup
One paid CLI gateway to 359 AI/vendor APIs — image, video, TTS, scraping, DB, auth, deploy.
- 230 Works with setup
Drives .NET, COBOL, VMware, SQL and Java/Python upgrades through the AWS Transform service
- 231 Works with setup
Works through a systematic workflow to debug hooks that misfire or misbehave.
- 232 Works with setup
Multi-agent GitHub PR review with confidence scoring and false-positive filtering
- 233 Works with setup
Generates Rails upgrade detection scripts and reports — solid for 7.2+, missing for the first 7.0→7.1 hop.
- 234 Works with setup
Structured git commit conventions, but its automation script is missing from the repo
- 235 Works with setup
Bedrock boto3/CLI snippets, IAM model ARNs, and exception-by-exception debugging
- 236 Works with setup
Darwin Godel Machine reference code for self-improving, self-evolving AI agents in Python
- 237 Works with setup
Reference for AgentDB advanced APIs: QUIC sync, hybrid search, distance metrics, sharding
- 238 Works with setup
Wraps Claude around the OpenAI Codex CLI for exec/resume coding runs — needs codex installed.
- 239 In test queue
Infers loop invariants and pre/postconditions for formal verification in Dafny or Coq.
- 240 In test queue
Adds MCP server integrations, connecting a custom tool or API via .mcp.json.
- 241 In test queue
Coordinate multi-agent teams with bounded roles, worktree isolation, verification gates.
- 242 In test queue
Acts as a DSA and LeetCode mentor with hints, code reviews, and mock interviews.
- 243 In test queue
Generates animated videos and motion graphics from natural language into a React project.
- 244 In test queue
Securely stores and retrieves API credentials for Claude, Gemini, and GitHub.
- 245 Tested · Didn't pass
Manages Jira issues and Confluence pages through the Atlassian MCP integration.
- 246 In test queue
Builds and debugs WhatsApp automation workflows using Kapso triggers and functions.
- 247 Tested · Didn't pass
CloudFormation patterns for Bedrock agents, knowledge bases, guardrails and data sources
- 248 Tested · Didn't pass
No SKILL.md exists for this slug in the agentops repo
- 249 Tested · Didn't pass
Adds or extends coding-agent support in cc-sdd by following its new-agent SOP.
- 250 In test queue
Convert Claude Code slash commands into portable Agent Skills.
- 251 In test queue
Delegates work to the Claude Code CLI, including headless and CI automation runs.
- 252 In test queue
Analyze patterns from another codebase and apply its best practices to yours.
- 253 In test queue
Deploys a Vibes app to Cloudflare Workers through the Deploy API.
- 254 Tested · Didn't pass
Executes Python via the ipybox MCP server for code-mode tool calling — requires the full ipybox plugin bundle.
- 255 Tested · Didn't pass
Viral code-intelligence MCP: real graph, fast queries — token claim measured inverted.
- 256 Tested · Works
TypeScript reference for PinMe Worker auth: create/verify/list Identity Platform users.
- 257 Tested · Works
Anthony Fu's opinionated ESLint, TypeScript, and monorepo conventions for JS/TS projects.
- 258 Tested · Works
Checklist-driven REST/GraphQL and TypeScript interface design with error and naming rules
- 259 Tested · Works
Operate KubeSphere FrontendIntegration resources and the frontend-forge extension via kubectl.
- 260 Tested · Works
Microsoft Azure WebJobs Extensions Authentication Events (.NET)
Reference for building Entra ID custom authentication event Azure Functions in .NET
- 261 Tested · Works
Adds a gated recharts chart to a PortalJS dataset showcase page.
- 262 Tested · Works
Scaffolds Angular 17+ standalone components, NgRx stores, and RxJS patterns with tests.
- 263 Tested · Works
SOP + code patterns for fixing browser automation failures on LinkedIn, X, SPAs, Shadow DOM.
- 264 Tested · Works
Fills CHANGELOG.md Unreleased with one linked line per commit since the last tag.
- 265 Tested · Works
Uncle Bob's Clean Code framework as a scored 0-10 rubric with six disciplines and reference deep-dives.
- 266 Tested · Works
Scaffolds VS Code .agent.md custom agent files with correct frontmatter, tools, and handoffs.
- 267 Tested · Works
Opens a Changesets-aware PR to main for the cc-wf-studio pnpm monorepo, never skipping the release changeset.
- 268 Tested · Works
Root-cause-first debug protocol that blocks guess-and-check patching after a failed fix
- 269 Tested · Works
Diagnosis-first Terraform/OpenTofu guidance with version-aware guardrails and reference docs.
- 270 Works with setup
Interactive 4-branch diagnostic that walks a developer through why their LLM won't call, or mis-calls, a tool.
- 271 Tested · Works
State-machine orchestration from issue intake to evidence-verified acceptance, never claiming done without proof.
- 272 Tested · Works
Detects your stack and wires lint/typecheck/test/format quality gates into .claude/settings.json.
- 273 Tested · Works
Executes the AGENTS.md build plan: plan, build one feature, verify, then update progress.
- 274 Works with setup
Answers codebase questions by grepping a pre-built knowledge-graph.json instead of re-reading the whole repo.
- 275 Works with setup
Non-interactive Vercel deploys with pre-build and post-deploy live-URL verification.
- 276 Tested · Works
Findings-only code review: hunt wide, prove every bug, no praise filler.
- 277 Tested · Works
Validator-clean REST/OpenAPI specs with a mandatory cross-endpoint consistency audit.
- 278 Tested · Works
Commit messages that match the diff, not just the format — measured, not assumed.
- 279 Tested · Works
Changelogs that trace to real commits — breaking changes never buried or dropped
- 280 Tested · Works
Flags speculative or unused new API surface before reviewing implementation quality.
- 281 Tested · Works
KubeSphere DevOps: Credentials
Manages repository, kubeconfig, and API token credentials in KubeSphere DevOps.
- 282 Works with setup
Analyzes a git diff or PR to summarize changes, affected components, and risks.
- 283 Works with setup
Scaffolds production-ready FastAPI projects with async patterns and error handling.
- 284 Works with setup
Finds gaps between documented intent and actual code, like undocumented access-control drift.
- 285 Tested · Works
Produces a file-cited technical debt report across nine dimensions with fix priorities.
- 286 Works with setup
Runs a version-aware release workflow across Node.js, Python, Rust, and other projects.
- 287 Works with setup
Drives interactive terminal sessions and REPLs while preserving state across calls.
- 288 Works with setup
Adds Vercel deployment capability to NanoClaw agent containers via the Vercel CLI.
- 289 Tested · Didn't pass
Routes a NanoClaw agent group to a local Ollama model via env var overrides.
- 290 Tested · Didn't pass
Reviews code across multiple quality dimensions before it merges into main.
- 291 In test queue
Set up the Dynatrace iOS SDK via Swift Package Manager with config and privacy opt-in.
- 292 In test queue
Guide ARM Cortex-M and RISC-V flashing, debugging, and RTT workflows via probe-rs.
- 293 Tested · Didn't pass
Routes 'use rp to...' requests to RepoPrompt's token-efficient codebase-exploration CLI.
- 294 In test queue
Explores an indexed codebase to explain architecture and trace execution flows.
- 295 In test queue
Review Go code against community style standards before a PR.
- 296 In test queue
Queries Google's AI Search mode for cited overviews on current topics and code.
- 297 In test queue
Upgrades Meta Quest apps to a newer Horizon OS SDK and replaces deprecated APIs.
- 298 In test queue
Swipe through every installed skill and keep, delete, or skip it one at a time.
- 299 In test queue
Generates JeecgBoot CRUD code for new modules and modifies existing fields.
- 300 In test queue
Generates JeecgBoot CRUD code and creates new modules with fields.
- 301 In test queue
Turn a feature request or migration into an implementation plan with clear ownership.
- 302 In test queue
Creates a new Langfuse integration or cookbook page for a tool, framework, or provider.
- 303 In test queue
Run the full math-modeling contest workflow from topic pick to final paper review.
- 304 In test queue
Implement Next.js server actions, parallel routes, intercepting routes, and streaming.
- 305 In test queue
Troubleshoots WhatsApp message delivery and webhook issues in Kapso via logs.
- 306 In test queue
Orients a new developer with a guided interview and codebase overview.
- 307 In test queue
Manages Paperclip AI agent companies, including tasks, hires, and heartbeats.
- 308 Tested · Didn't pass
Blocks inline multi-file plans and routes them through a local browser annotation UI instead.
- 309 In test queue
Reviews code against markdown requirements or specs and applies focused fixes.
- 310 In test queue
Mine session history and transcripts for recurring workflows worth turning into skills.
- 311 In test queue
Publishes static page or React artifacts to SourceCraft Sites under a dated layout.
- 312 In test queue
Require real evidence code works before approving a verify pass, not just green tests.
- 313 In test queue
Requests and manages a Superior Trade API key for authenticating with api.superior.trade.
- 314 In test queue
Scans a codebase for technical debt, scores severity, and produces a remediation plan.
- 315 In test queue
Plans, splits, and delegates a coding task across subagents, then integrates and verifies.
- 316 In test queue
Bumps version numbers across the Claude Code Handbook monorepo during release prep.
- 317 In test queue
Get current weather, temperature, and forecasts for any location.
- 318 Tested · Works
Alpaca Broker Account Onboarding
Open and manage Alpaca brokerage accounts: KYC, agreements, documents, status lifecycle.
- 319 Works with setup
Playbooks that tell Claude which Langfuse MCP tool to call first when debugging LLM traces.
- 320 Tested · Works
Developer guide to implement compact memory/context compaction in Claude agents.
- 321 Tested · Works
Runs Codex/OpenCode/Gemini/Claude in parallel for independent second opinions and 2-specialist code review.
- 322 Tested · Works
CesiumJS camera reference: flyTo/lookAt/setView with altitude-pitch tables and navigation-constraint gotchas.
- 323 Works with setup
Turns a diff into a ranked, graph-validated review brief with an anti-hallucination judgment loop.
- 324 Tested · Works
Strict senior-engineer persona: mode triggers, spec-plan-TDD-verify loop, security floor, tiered CI gates.
- 325 Works with setup
Reference for the neo4j-agent-memory Python package: three-layer graph memory (short-term/long-term/reasoning) for AI agents.
- 326 Tested · Works
A pre-refactor checklist that catches untested behavior before you 'clean up' code.
- 327 Tested · Works
Configures Better Route's audit-log middleware so idempotency keys get hashed and logger failures never break the API response.
- 328 Tested · Works
Read-only, sub-agent-forked code health scanner with a fixed severity-graded report format.
- 329 Tested · Works
Python component framework for type-safe HTML/XML/RSS generation.
- 330 Tested · Works
Builds and verifies Pyxel games using real headless validate/run evidence, not vibes.
- 331 Tested · Works
Writes and validates MikroTik RouterOS /app YAML — a docker-compose lookalike with real differences
- 332 Tested · Works
Builds a spec-correct A2A Agent Card JSON by fetching the live protocol schema first.
- 333 Tested · Works
Adds a new cybersecurity tool to this specific installer's modules, config, and validators.
- 334 Tested · Works
File-based per-project chat (channels + DMs) so multiple AI coding agent instances can coordinate.
- 335 Tested · Works
Decision-tree methodology for scoping LLM pipeline projects: task-model fit, staged architecture, cost math.
- 336 Tested · Works
Analyze M1 Module For Migration
Decides migrate/replace/skip for a Magento 1 module before an M2 port
- 337 Tested · Works
Angular Enterprise Architecture
Enforces lazy-feature isolation and a one-way dependency graph in Angular apps, with an ESLint boundaries config to lock it in.
- 338 Tested · Works
Exact ArcGIS JS SDK 3D layer syntax: SceneLayer, PointCloudLayer, VoxelLayer, DimensionLayer, glTF.
- 339 Tested · Works
AWS CloudFormation ElastiCache
CloudFormation patterns for private, encrypted, Multi-AZ ElastiCache Redis/Memcached tiers.
- 340 Tested · Works
Generates AWS architecture/security/network/data-flow diagrams from an infrastructure JSON snapshot.
- 341 Tested · Works
Turns a requirement doc into dependency-ordered module plans, a checklist.md, then executes them with subagents.
- 342 Tested · Works
Governed CLI for Bitrix24 REST integrations with plan-then-execute writes and audit logging.
- 343 Works with setup
Hard-gates code writing behind an approved design doc and a bite-sized, placeholder-free implementation plan.
- 344 Tested · Works
Dense SimBiology API/rules reference that prevents real ambiguous-species and deprecated-property errors.
- 345 Tested · Works
Scores a repo's Docker/Kubernetes readiness 0-12 and hands off to a Dockerfile generator.
- 346 Tested · Works
Disciplined LSP-vs-rg-vs-fuzzy search precedence with position-anchoring and forced disclosure of tool substitutions.
- 347 Tested · Works
Seemann's engineering heuristics (complexity limits, TDD, API design, STRIDE) routed to the exact file a task needs.
- 348 Tested · Works
Concrete clean-code ruleset: 2-param max, no error suppression, early returns, YAGNI.
- 349 Tested · Works
Splits mixed diffs into logically-scoped, Conventional-Commits-formatted commits.
- 350 Tested · Works
Fetches live, sourced library docs via the Context7 API instead of relying on training data.
- 351 Tested · Works
Diagnose and fix Coolify deployments, WordPress issues, and SSL certs via CLI and API.
- 352 Tested · Works
Builds valid Vapi voice-agent JSON payloads with vetted defaults, no invented IDs.
- 353 Tested · Works
Bootstraps a business-domain (not folder-based) doc system for AI coding agents, with a machine-checked staleness gate.
- 354 Tested · Works
Runs the DocumenterVitepress.jl local edit-rebuild-preview loop for Julia docs sites correctly.
- 355 Tested · Works
C#/EPLAN P8 scripting patterns plus a live 55k-vector RAG over the full docs
- 356 Tested · Works
Ten project-specific ESLint fix patterns (no-void, sonarjs/deprecation, lodash bans, jest-dom) with before/after code.
- 357 Tested · Works
Makes diffs match the file's existing idiom instead of your default habits.
- 358 Tested · Works
Filament PHP v5 Coding Standards
Enforces correct Filament v5 API (recordActions, Heroicon enum, Operation enum) over stale v3/v4 patterns.
- 359 Tested · Works
Keeps GitHub Actions workflows on current action versions with least-privilege security defaults.
- 360 Tested · Works
Official Glue IDL guide: model once, generate TS/Python/Rust/Go/OpenAPI/Protobuf.
- 361 Tested · Works
Composes a paste-ready /goal block that loops plan review-and-fix until APPROVED.
- 362 Tested · Works
Production-grade GitHub Actions CI for Go: test, lint, SAST, releases, dependency bots.
- 363 Tested · Works
Cuts a gumroad-cli release with the repo's actual Go-date tag scheme (v0.YYYYMMDD.N), validated end to end against the live repo.
- 364 Tested · Works
Builder-style HXMusic/HXSound API reference for Android background music and sound effects
- 365 Tested · Works
Wire production Inngest agent evals: direct/deferred scoring, sessions, and A/B experiments.
- 366 Tested · Works
Enterprise Java checklists that force a cursor/transaction/mapper fix to actually enforce the invariant, not just mention it.
- 367 Tested · Works
Production-readiness audit for vibe-coded apps: scored scorecard plus fix list
- 368 Tested · Works
Turns a vague engineering ask into a runnable /goal contract with constraints, done-when, and stop rules.
- 369 Tested · Works
Supply-chain checklist that verifies a package via real registry APIs before you add it, and pushes you to inline instead.
- 370 Tested · Works
Wires the correct minimal @mobile-reality/mdma-* package set for React and LLM-streaming integrations.
- 371 Tested · Works
Persistent .memory-bank/ project memory, TDD/Clean-Architecture rules, and a 25+ command dev toolkit.
- 372 Tested · Works
Interactively scaffolds a new sub-project into a MoxPP (Premake5+Conan2) C++ workspace.
- 373 Tested · Works
PR/release governance playbook for this repo: change classification, exact gate order, and cited non-negotiables.
- 374 Works with setup
Documents claude-night-market's own config axes and guards — accurate but repo-bound, not portable.
- 375 Tested · Works
Node.js XML-RPC patterns for syncing Odoo stock/sales/purchase data to Sheets, DBs, or Railway.
- 376 Tested · Works
Codebase-expert prompt for Kye Gomez's OpenMythos looped-transformer PyTorch repo.
- 377 Tested · Works
Senior-level architecture review of one React feature with file:line findings and fixes.
- 378 Tested · Works
Blocks sycophantic PR-review capitulation and forces evidence before implementing 'best practice' suggestions.
- 379 Tested · Works
Turns a single-user app into multi-tenant SaaS: auth, org-scoped RLS, RBAC, billing, invites in order.
- 380 Tested · Works
Docs-first iOS Screen Time (FamilyControls/ManagedSettings/DeviceActivity) implementation reference.
- 381 Tested · Works
Read-only OWASP Top 10 guardrail for Python/Django/Flask/FastAPI code generation - no scripts, just rules.
- 382 Tested · Works
Picks the cheapest Sendmux attachment path (file_path or presigned URL) instead of burning context on base64.
- 383 Tested · Works
Governed Sentry metrics and gen_ai tracing, enforced by a real AST CI gate.
- 384 Tested · Works
Registers custom Simulink block libraries so the agent prefers them over built-in blocks.
- 385 Tested · Works
Installs a two-tier codebase index (CLAUDE.md file map + per-directory YAML) with a hook that enforces upkeep.
- 386 Tested · Works
SG Deceptive Reachability Auditor
Composes AWS security-group references into a reachability graph to find the lateral-movement path no single rule reveals.
- 387 Tested · Works
Ports Shadertoy GLSL shaders into Flutter .frag files that actually compile and load.
- 388 Tested · Works
Rigid phase-by-phase plan executor enforcing tests, linters, and DoD gates.
- 389 Tested · Works
Doc-grounded code generation for the TinyRobot Vue AI-chat component library.
- 390 Tested · Works
CLI wrapper for the Unoplat Code Confluence app: repo onboarding and AGENTS.md generation with auto-PR.
- 391 Tested · Works
C++ recipes for generating UE5 Blueprint and Widget Blueprint .uasset files programmatically.
- 392 Tested · Works
Reference tables + wrapper-pattern workflow for moving classic ABAP to ABAP Cloud Tier 1.
- 393 Tested · Works
Unity Addressables cookbook: handle-lifecycle discipline, label loading, and the IsValid()-before-release rule most devs skip.
- 394 Works with setup
Audits an existing codebase and retrofits agent-starter's hooks, lint configs, and patterns, tier by tier, with approval gates.
- 395 Tested · Works
Byte-verified copy of your canonical AGENTS.md into a repo, plus a structured repo-specific tail
- 396 Tested · Works
Writes a CLAUDE.md/AGENTS.md that lists only verified commands and real gotchas
- 397 Tested · Works
Keeps a repo decision journal agents must read before changing covered code
- 398 Tested · Works
Design agent/CLI tool schemas, naming, errors, and idempotency for LLM callers.
- 399 Tested · Works
End-to-end Salesforce Agentforce agent lifecycle: design, Agent Script, validate, deploy, publish.
- 400 Tested · Works
Provider-neutral agent harness design: loops, tools, permissions, evals
- 401 Works with setup
Reference docs for AceDataCloud's paid unified LLM gateway (GPT/Claude/Gemini/Grok/50+ models) -- auth guide lives in a sibling folder the verbatim install leaves behind.
- 402 Tested · Works
Bootstraps a multi-agent dev/QA team with sprint plans and cross-chat handoff docs.
- 403 Tested · Works
Reference patterns for versioning, deprecating, and migrating REST APIs without breaking clients.
- 404 Tested · Works
Upload local screenshots/videos to GitHub and get markdown-ready asset URLs
- 405 Tested · Works
Classic AUTOSAR BSW expert: config, ARXML/RTE debug, COM stack, boot/NVM, UDS
- 406 Tested · Works
Production CloudFormation patterns for CloudFront: origins, caching, WAF, and security headers.
- 407 Tested · Works
Layered backend patterns: errors, logging, caching, and API conventions across 6 languages.
- 408 Tested · Works
Structured 6-step BLE GATT debug workflow separating radio, discovery, attributes, security, and throughput
- 409 Tested · Works
Real Rust linter that catches code/docs drift via <block affects=...> comment tags -- verified with the actual built binary, not just reading the source.
- 410 Tested · Works
Scaffold and extend local-first TinyBase JS/TS apps with correct persistence and sync wiring
- 411 Tested · Works
Build Caido web-security plugins with exact frontend/backend/workflow SDK signatures
- 412 Tested · Works
Reverse-engineered Claude Code built-in agent configs (tools, model routing, omitClaudeMd) turned into a design checklist for your own subagents.
- 413 Tested · Works
Production n8n patterns for multi-stage LLM extraction, classification, and scoring chains.
- 414 Tested · Works
Split a ticket into requirement blocks and grade each done / partial / not-done / not-verifiable against the code
- 415 Tested · Works
Leaf-vs-core decision framework for how much AI/review rigor a code change needs.
- 416 Tested · Works
One-shot migrate code and prompts from Sonnet/Opus 4.1 to Opus 4.5
- 417 Tested · Works
Repo-embedded debugging map for ClawBox's chat SSE streaming across the React frontend and Bun bridge
- 418 Tested · Works
Converts a Lanhu HTML export into Vue or React that reuses your components
- 419 Tested · Works
Refactors recently-touched code for clarity while proving behavior is unchanged -- kills nested ternaries and magic numbers.
- 420 Tested · Works
Eleven task modes with checklists, picked before non-trivial coding work
- 421 Tested · Works
Systematic, changelog-first playbook for safe Composer dependency updates.
- 422 Tested · Works
Simulates named expert panels (Fowler, Hickey, Fielding...) to reason through a design or architecture question.
- 423 Tested · Works
Installs Snowflake CLI and Cortex Code CLI, OS-aware, with MCP-conflict guard.
- 424 Tested · Works
Craft CMS 5 content architecture with a reserved-handle gate
- 425 Tested · Works
Scan a codebase for REST, GraphQL, gRPC and OpenAPI definitions
- 426 Tested · Works
Five-phase kickoff workflow (research, design, plan, scaffold, document) for turning a tool idea into a shippable MVP.
- 427 Tested · Works
Genkit Dart SDK reference that actually compiles — verified against real pub.dev packages.
- 428 Tested · Works
Fetches the real @dile/ui component docs before you write dile-* markup
- 429 Tested · Works
Structured .NET PR review with adversarial mode and heuristic risk-grep scripts.
- 430 Tested · Works
Adversarial pre-mortem gate for risky diffs, migrations, and 'this is safe' claims
- 431 Tested · Works
Repairs eBPF verifier rejections from the raw verbose load log, proof-first
- 432 Tested · Works
Opinionated Elixir/Phoenix/OTP architecture playbook that outputs a full ADR-and-guardrail documentation package, not just code.
- 433 Tested · Works
Fable-ish Verification Discipline
Risk-proportional verify-before-you-claim-done workflow for coding tasks, ported from a Codex plugin.
- 434 Tested · Works
Runs the full Atlassian Forge app workflow: create, deploy, install, with real scripts.
- 435 Tested · Works
Structured, AI-driven security review of a diff/PR against 18 Fortify vulnerability checklists.
- 436 Tested · Works
Use the gh CLI instead of web-fetching GitHub issue/PR/repo/discussion links.
- 437 Tested · Works
One engineer's Git branch/commit/push rulebook - naming convention, JetBrains DontCommit guard, Chinese commit defaults
- 438 Tested · Works
GitHub Actions, PR, issue, and release workflow reference for Claude Code plugin repos.
- 439 Tested · Works
GitLab issue/MR fetcher, executive-summary and code-review generator, and auto-resolve-to-MR workflow via a real CLI.
- 440 Tested · Works
Go slice/map internals and container-package selection guidance, with idiomatic generics.
- 441 Works with setup
Login, profile switching, and keyring-credential diagnosis for the official Halo CMS CLI.
- 442 Tested · Works
Scaffolds AGENTS.md, harness scripts, and CI so agents run repos reliably
- 443 Tested · Works
Pushes frontend builds toward native HTML/CSS (details, popover, :has(), scroll-driven animation) before reaching for JS.
- 444 Tested · Works
Turn a codebase into a single self-contained HTML portal that works offline on file:// wit
- 445 Works with setup
Builds a 4-screen Socratic React teaching artifact from '/aha [topic]' — designed for claude.ai's artifact sandbox, not Claude Code.
- 446 Tested · Works
Designs NATS JetStream streams, subjects and consumers, with working Go/JS/Python code.
- 447 Tested · Works
Idiomatic Keras3 patterns for R: Sequential/Functional/Subclassing APIs, multi-backend, callbacks.
- 448 Tested · Works
Scaffolds Laravel REST APIs with invokable controllers, FormRequest DTOs, Action classes, and ULIDs.
- 449 Tested · Works
Wizard-driven setup of an isolated Lima VM for running AI coding agents in YOLO mode.
- 450 Tested · Works
PEFT/LoRA and QLoRA config recipes for fine-tuning LLMs on limited GPU memory.
- 451 Tested · Works
Alpine.js reactive component patterns tuned for Magento 2 / Hyvä themes
- 452 Tested · Works
One command that scaffolds a full Claude Code agent team, a single agent, or a single skill.
- 453 Tested · Works
Dense NATS/JetStream reference covering streams, consumers, KV, security, clustering, and a common-mistakes table.
- 454 Works with setup
C-suite/compliance/PM/finance reference templates for a single solo-founder's Cloudflare-Workers stack.
- 455 Tested · Works
Stops PowerShell scripts from breaking on quoted paths, Invoke-Expression, and Start-Process quirks.
- 456 Tested · Works
A 16-rule senior-engineer posture: plan first, atomic writes, typed errors, schema+queries+indexes as one diff.
- 457 Tested · Works
Splits oversized SKILL.md files into linked reference files without rewriting content.
- 458 Works with setup
Publish Release (moi-computer)
npm release runbook: local verify, tag push, gated CI publish, then restore
- 459 Tested · Works
Recovers stuck Pulumi stacks, fixes state drift, and sets up non-interactive CI/CD with exact flags.
- 460 Tested · Works
Opinionated Python patterns for CLI tools, async, FastAPI, jobs and observability.
- 461 Tested · Works
Decision-tree entrypoint into 25+ Quarkus reference modules with copy-ready, current APIs.
- 462 Tested · Works
Pressure-tests an implementation plan with adaptive multi-perspective subagents before you write code.
- 463 Tested · Works
Diff-scoped review orchestrator with real automated gates, severity/confidence-tagged findings, and a mandatory triage menu.
- 464 Tested · Works
Claude plans, Codex CLI attacks the plan and builds it, Claude runs the proof
- 465 Works with setup
CLI to inspect and control ROS/ROS2 robots over rosbridge WebSocket - topics, services, nodes, actions.
- 466 Tested · Works
Sets up branch-isolated Supabase/Vercel/GitHub Actions preview databases for AFK agent workflows.
- 467 Tested · Works
In-session LLM SAST triage with adversarial verify and schema-checked findings
- 468 Tested · Works
Scaffolds a Vapor or Hummingbird backend with Fluent, JWT/Apple auth and Docker deploy, decision matrix included.
- 469 Tested · Works
Version-aware sing-box client config generator with hard rules against deprecated/legacy field misuse.
- 470 Tested · Works
Scaffolds and validates new Claude skills with enforced structure and metadata.
- 471 Tested · Works
SOLID refactoring with restraint — measurably less abstraction bloat
- 472 Tested · Works
Turns raw SpecStory session logs into a standup-ready summary with outcome emoji and a patterns section.
- 473 Tested · Works
Streaming 4K image CLI for sub2api endpoints that dodges Cloudflare 524 timeouts
- 474 Tested · Works
Swift 6 data-race diagnostics, actor isolation, callback-to-async migration
- 475 Tested · Works
Runtime-level SwiftUI review for macOS: observation, concurrency, AppKit
- 476 Tested · Works
React to Doctrine entity lifecycle in Symfony with ORM 3 attribute listeners
- 477 Tested · Works
Three.js recipe for layered-PNG 2.5D parallax cards with mouse + scroll drift.
- 478 Tested · Works
Current-API reference for Three.js WebGPU renderer and TSL node shaders
- 479 Tested · Works
Yandex Games SDK integration (ads, saves, leaderboards) and moderation checklist for HTML5, Godot, and Unity games -- always re-fetches live rules first.
- 480 Works with setup
Turns any YouTube channel into a searchable Claude Code advisor skill.
- 481 Tested · Works
.NET Clean Architecture Scaffolder
Scaffolds a compile-ready .NET Clean Architecture solution across four layers
- 482 Works with setup
LSP setup diagnostic and fixer for Rust/TS/Ruby, but only as a full plugin.
- 483 Tested · Works
Three isolated subagent personas (Saboteur, New Hire, Security Auditor) hostile-review your diff in parallel.
- 484 Tested · Works
Researches and drafts a multi-agent system architecture blueprint from a workflow description.
- 485 Tested · Works
Hardening playbook for native agents shipped onto machines you don't control
- 486 Tested · Works
Two-stage recipient whitelist check + symlink-safe worktreePath validation for multi-agent SendMessage.
- 487 Tested · Works
Turns an Agent Build Spec into a runnable Python agent: loop, permissions, logs
- 488 Tested · Works
Structured P0-P3 diff review: correctness, security, tests, perf, before merge.
- 489 Tested · Works
Anti-pattern guardrails (no silent assumptions, no scope creep) for Agile V coding agents
- 490 Tested · Works
CLI-first integration layer for Agnes AI text, image, and video generation.
- 491 Tested · Works
Sets up AGENTS.md, packaged skills, and MCP tooling so AI coding agents understand and self-improve on a codebase.
- 492 Tested · Works
Turns source code into structured API docs with a completeness-enforcing template
- 493 Tested · Works
Reference TypeScript pattern for wrapping an external HTTP API: env secrets, bounded retry
- 494 Tested · Works
Scaffolds a typed API client with retry, rate limiting, and auth from a spec
- 495 Tested · Works
Swift 6 concurrency, SwiftUI, Metal, and RealityKit/visionOS engineering standards for Apple platform code.
- 496 Tested · Works
Named vocabulary and consequence-framing for spotting god objects, tight coupling, and shared-DB coupling.
- 497 Works with setup
Maps Delphi classes to relational tables with TMS Aurelius ORM attributes, including lifetime-safety rules.
- 498 Tested · Works
Baidu Maps BMapGL reference pack: 24 API docs for overlays, layers, routing
- 499 Tested · Works
Bank of Georgia payment API reference: OAuth, orders, receipts and refunds
- 500 Tested · Works
Structured reproduce-then-fix workflow for red CI pipelines and containers that won't start.
- 501 Tested · Works
Scans AI-written code for 16 structural-slop patterns without changing behavior
- 502 Tested · Works
Interactive code walkthrough that runs each section and shows real output
- 503 Tested · Works
Groups pending git changes into logical Conventional Commits, with a secrets guard and an approval pause before each commit.
- 504 Tested · Works
Design the API contract before the code: versioning, idempotency, cursor paging
- 505 Tested · Works
Teaches the exact @convex-dev/agent API for threads, streaming, and tool-calling agents.
- 506 Tested · Works
Reference guide for generative art: flow fields, L-systems, shaders, particle text, polar art.
- 507 Tested · Works
Cross-Cutting Checklist Generator
Builds a project checklist so scattered concerns never get half-updated
- 508 Tested · Works
Bug-triage persona: ranked root-cause hypotheses with confirm-steps, minimal fix, regression note
- 509 Works with setup
Spec-backed workflow for building and packaging ONES Apps (.opkx) via the ones CLI.
- 510 Works with setup
AI DevKit planning phase: turns requirements/design/testing docs into a traceable, milestone-based task plan.
- 511 Works with setup
Wires up DevExpress's DxAIChat Blazor component end to end — BYOK OpenAI/Azure/Ollama chat UI, v26.1 API only.
- 512 Works with setup
Create, edit, and validate Dify workflow DSL YAML from the command line
- 513 Tested · Works
Divi 5.6 CSS compatibility reference: specificity fixes, units, breakpoints, bug history
- 514 Tested · Works
Doctor: Dashi Telegram Plugin Migration
Read-only doctor for migrating a Telegram bot to the dashi channel plugin
- 515 Works with setup
Derivation-first Dynare .mod writer/debugger with strict timing and naming rules
- 516 Tested · Works
Decision tables for eBPF: hooks, maps, kernel minimums, verifier fixes
- 517 Tested · Works
Frontend API client discipline: typed boundaries, error mapping, auth refresh
- 518 Works with setup
Orchestrator that drives an 11-phase Flutter/Flame game pipeline from idea to app-store submission.
- 519 Tested · Works
Turns branch diff into a structured PR title and description with per-module test cases
- 520 Tested · Works
Catches Go's classic range/break/defer-in-loop footguns during code review.
- 521 Tested · Works
Enforces GoFrame v2 ORM conventions: DO objects, auto soft-delete and time
- 522 Tested · Works
Go database/sql, sqlx, pgx best practices — parameterization, transactions, pooling, NULLs.
- 523 Tested · Works
Portable exhaustive multi-agent audit/review protocol: roles, adversarial verify, loop-until-dry.
- 524 Tested · Works
Adversarial one-question-at-a-time interrogation of a plan, grounded in the actual codebase.
- 525 Tested · Works
Helm chart authoring conventions: registry overrides, labels, RBAC, templating
- 526 Tested · Works
Forces spec, RED/GREEN evidence and adversarial diff review before code ships
- 527 Tested · Works
Align-then-build workflow for runnable React/Vite hero landing pages
- 528 Tested · Works
IPSW Apple Reverse Engineering Toolkit
Drive the ipsw CLI to reverse-engineer iOS/macOS: DSC, kernel, entitlements, sandbox
- 529 Tested · Works
Audited sprint loop for large ambiguous specs — proof obligations over checked boxes.
- 530 Tested · Works
Spring Boot cache setup, review checklist and eviction patterns for Redis and Caffeine.
- 531 Tested · Works
Scores any spec/plan/PRD for overlaps and gaps (Mutually Exclusive, Collectively Exhaustive) and asks before fixing.
- 532 Tested · Works
Picks the right memory layer (file/vector/graph/temporal-graph) per requirement instead of one-size-fits-all
- 533 Tested · Works
NanoResearch Experiment Skeleton
Turns an experiment blueprint JSON into a runnable multi-module ML training/eval/ablation skeleton
- 534 Works with setup
Generates AGENT_MAP.md/CLAUDE.md/.mcp.json for a NixOS or Home Manager repo, fact-traced via the nix-agent MCP server.
- 535 Tested · Works
Auto-updated NixOS/Nix manual, nix.dev tutorials, and Nix Pills wired into a topic-to-file routing table.
- 536 Tested · Works
Odoo-specific PR review: ORM/sudo/security/performance checklist plus the official Odoo 19 style guide.
- 537 Tested · Works
Database design specialist: schema layers, 3NF/BASE tradeoffs, indexing, and vector/RAG retrieval architecture.
- 538 Works with setup
Monorepo task automation specialist for mise-based lint/test/build/release pipelines.
- 539 Tested · Works
Set up and troubleshoot the OpenTag CLI that links Slack/GitHub to coding agents
- 540 Tested · Works
Config reference for CrestApps' Orchard Core A2A Client feature: connections, auth types, and profile wiring.
- 541 Works with setup
Recommends one payment provider (Stripe/MP/Wompi/Lemon Squeezy) with real fee math, no vendor talk.
- 542 Tested · Works
Whole-repo over-engineering audit: a ranked delete/simplify/replace list with net-line count
- 543 Tested · Works
Apple's full Swift book (6.4 beta) as an offline language-reference skill
- 544 Tested · Works
Runs the Railway CLI for you: status, logs, deploy, vars, health checks.
- 545 Tested · Works
Enforces FAIL FAST Rust: propagate every error, thiserror/anyhow, edition 2024
- 546 Works with setup
Turns GitHub issues/PRs into a signed, queryable knowledge graph of your work
- 547 Tested · Works
Walks step-by-step through defining, writing, testing, and shipping a Claude Skill, with the real platform constraints baked in.
- 548 Works with setup
Pharo/Smalltalk error-diagnosis and Transcript-logging workflow for AI coding agents.
- 549 Works with setup
Walks you through connecting the Stitch MCP server and installing the stitch-kit plugin.
- 550 Works with setup
Turns a git diff into a structured, linked HTML PR summary — but needs `npm install` first, undocumented.
- 551 Tested · Works
Swift Modern Architecture Skill
Forces Swift 6/iOS 18+ idioms (Observable, SwiftData, async/await) over legacy patterns.
- 552 Tested · Works
SfAccordion for .NET MAUI: correct NuGet package, XAML namespace and events
- 553 Works with setup
Turns the conversation you already had into a repo-aware spec in .plan/
- 554 Tested · Works
Explore-first planner that turns vague briefs into one decision-complete plan.
- 555 Tested · Works
Roast and strip AI code slop: narrator comments, vacuous tests, over-abstraction
- 556 Works with setup
Teaches Claude to never echo secrets — validates .env vars through the Varlock CLI instead of cat/echo.
- 557 Tested · Works
One-pass Docker + compose + CI/CD generator with a real deployment audit script
- 558 Works with setup
Intent-routed Exa MCP search with quick/deep/code/docs/debug/news/compare modes and forced citations.
- 559 Works with setup
Installs an AGENT_NAME-isolated Makefile + script toolkit for CLI-only Xcode builds.
- 560 Tested · Works
Converts Android XML layouts to idiomatic Jetpack Compose with modifier-order rules.
- 561 Works with setup
Multi-agent epic-to-PR workflow: 4-phase execution, tiered models, TDD gates
- 562 Tested · Works
Reference guide for authoring and debugging Claude Code SKILL.md files.
- 563 Works with setup
Delegate coding tasks to Google's Antigravity CLI with a stall-safe headless runner.
- 564 Tested · Works
Generates runnable akquant quant-strategy code: hooks, backtest, risk, optimizat
- 565 Tested · Works
Angular Material M3 theming and component guidance, correctly steers off the deprecated M2 API
- 566 Tested · Works
Systematic REST/GraphQL API review with P0-P3 prioritized findings, effort estimates, and a launch checklist.
- 567 Tested · Works
Subprocess-safe API/CLI patterns: no shell=True, whitelisted commands, retry backoff, token validation.
- 568 Works with setup
API Integrations (Falcon Foundry)
Downloads, auto-fixes, and registers vendor OpenAPI specs as CrowdStrike Falcon Foundry integrations.
- 569 Works with setup
Exact `asc` CLI invocations for build processing status, next build numbers, and expiring old builds.
- 570 Works with setup
Senior-architect review persona: 9-point structural checklist, YAGNI/astronaut-architecture guardrails.
- 571 Tested · Works
Turns a PR diff into a before/after architecture doc for PMs, with an honest risk section.
- 572 Works with setup
Weighted 7-dimension code/design quality scorer with a pass/fail gate — needs the full ork plugin plus memory MCP.
- 573 Tested · Works
CloudFormation snippets for CloudWatch alarms, dashboards, log groups, and metric filters, with deploy steps.
- 574 Works with setup
Lets one Claude Code session query a connected peer session for live, full-context answers.
- 575 Works with setup
Cited, multi-source research with a mechanical grounding checker — but ships wired into a whole sibling-skill suite.
- 576 Tested · Works
Comment on a served HTML page; Claude fixes it and morphs the tab, no reload
- 577 Works with setup
Cross-repo research workflow for self-hosted cix workspaces: which repos, which code, what changes.
- 578 Works with setup
Real, working Claude<->Codex audit-handoff pipeline with durable run tracking, but the skill scripts are dead on arrival without a separately-installed bridge runtime
- 579 Tested · Works
Diffs package.json, requirements.txt, go.mod and friends against your CLAUDE.md
- 580 Tested · Works
Two-mode tech-debt scanner: DRY/naming/length sweep plus pre-release repo audit
- 581 Tested · Works
Turns a design doc into a phase-by-phase implementation plan with cost estimates and an approval gate.
- 582 Works with setup
Semantic search, grep, and call-graph tracing across CodeAlive-indexed repos.
- 583 Works with setup
Evidence-first GitHub PR, CI-failure and fork-sync review CLI, no key needed
- 584 Tested · Works
CometChat Native Bare Patterns
CometChat integration recipe for bare React Native CLI: pods, permissions, Apple
- 585 Works with setup
Shapes a coding task into a five-section Cursor/Composer delegation prompt
- 586 Works with setup
D2 Diagram Generator for Azure
Turns Terraform Azure infra into icon-accurate D2 architecture diagrams via a mandatory scope questionnaire and live icon lookup.
- 587 Tested · Works
Reference-backed LangGraph 1.0 / LangChain 1.x agent-building guide with 11 chapters
- 588 Works with setup
Dispatch-verify-synthesize orchestration across Claude/Codex/OpenCode — but needs a separate agent-mux CLI to actually run.
- 589 Tested · Works
Replaces ad-hoc trial-and-error fixing with an ANALYZE-PLAN-MUTATE-VERIFY-SCORE-ARCHIVE loop logged to .evolution/log.json.
- 590 Works with setup
A real, well-grounded architecture-review rubric for the jup CLI's own codebase — but it only means anything if you're literally working inside the jup repo.
- 591 Works with setup
Detects Laravel/Next.js/React and plans a modular-architecture migration — but hands off to plugins not in the box.
- 592 Works with setup
Jekyll and Actions GitHub Pages setup, hardcoded to the SenolIsci/mykg repo
- 593 Tested · Works
PHP 8.1-8.5 modernization workflow with a real discovery script, checklist, and hard guardrails.
- 594 Tested · Works
Multi-angle production-readiness audit with file:line evidence and no hedging
- 595 Tested · Works
Rewrites vague Claude Code requests into scoped, stack-aware prompts before execution.
- 596 Works with setup
Gated Socratic feature-design interview, but requires the whole quantum-loop plugin, not just this folder
- 597 Tested · Works
Generates a full Rust architecture doc package: ADRs, guardrails, domain model, Director/Implementor handoff.
- 598 Tested · Works
Drives the ServiceNow Fluent CLI: authoritative explain docs and live instance queries
- 599 Works with setup
Browse SmartCMP cloud resources, inspect VM detail, and run confirmed power operations.
- 600 Tested · Works
Curated engineering rules that Claude consults before it writes any code
- 601 Works with setup
Drives Abaqus FEA from an agent via sim-cli; needs your own licensed install
- 602 Works with setup
Checks an IronLint adapter against the harness contract it was written for
- 603 Tested · Works
Reference guide for writing Claude Code subagent .md files: frontmatter, triggering examples, system prompts.
- 604 Works with setup
CLI to search, recover, and analyze Claude Code / AI Studio / Gemini CLI session history.
- 605 Tested · Works
Analysis Pipeline (Greenfield)
Reverse-engineer code into clean-room behavioral specs with provenance
- 606 Works with setup
APK triage workflow: inventory DEX and native payloads, then decompile with jadx
- 607 Works with setup
Runs real static-analysis tooling and ranks refactor hotspots by churn x complexity.
- 608 Works with setup
Audits a codebase for shallow/coupled modules and writes a deep-module refactor PRD.
- 609 Works with setup
Router skill that loads iOS/Swift skill packs and a safe build/test workflow
- 610 Works with setup
Cloud build/sign/publish pipeline for Swift apps via a proprietary signing service, no local Xcode needed.
- 611 Works with setup
Bumps the FutureSearch SDK version across all 11 config/lockfile locations in one pass
- 612 Tested · Works
House C# 12 / .NET 8 rules: Result pattern, EF query hygiene, async discipline
- 613 Works with setup
Pulls historical debugging fixes from ChatCrystal's memory before you diagnose a new failure.
- 614 Works with setup
Self-hosted GitLab MR reviewer: deterministic lint plus inline comments
- 615 Tested · Works
WHY-first code analysis with rejected alternatives and boundary conditions
- 616 Works with setup
Turns a constraints/NFR spec into a deterministic, cost-audited architecture pattern selection.
- 617 Tested · Works
Component Patterns (Prismatic)
Prismatic custom component conventions for the spectral TypeScript SDK
- 618 Works with setup
ControlFlow Executability Verifier
Simulates a saved plan's first tasks cold-start to catch vague phases before coding starts.
- 619 Works with setup
Executes a Carmack Council plan task-by-task, loading each expert's reference doc first.
- 620 Tested · Works
Scraped LangChain Deep Agents docs for backends, subagents, and human-in-the-loop config.
- 621 Works with setup
SDLC phase router for the ai-devkit CLI — requirements through review, gated by a live lint/skill-list check.
- 622 Tested · Works
Generates GitHub Actions CI/CD YAML with security and performance best practices baked in.
- 623 Works with setup
Persistent godot-mcp + AI Bridge inspect/edit/run/debug loop for Godot 4.x
- 624 Tested · Works
Golang Fullstack Best Practices
Rule-backed Go + GORM + PostgreSQL code review across 9 domains, 89 rules
- 625 Works with setup
Thin Claude Code entrypoint into the Hyper-Waterfall task-governance framework and CLI.
- 626 Works with setup
Project-internal engineering rules for the Lagune codebase (toolchain, conventions, build)
- 627 Works with setup
Multi-Model Adversarial Review
Runs claude/codex/gemini as independent reviewers, has them cross-critique, then synthesizes one deduped report
- 628 Tested · Works
Migrate n8n/Make workflows to Claude Code with node mapping + security scan
- 629 Works with setup
Root-cause debugging that enforces a red-then-green regression test on every fix.
- 630 Works with setup
Orchestrates one TDD iteration end-to-end: regression baseline, task decomposition, evidence tracking, wrap-up.
- 631 Works with setup
Read-only health audit for speq spec libraries with gated fixes
- 632 Tested · Works
Linus-style code-review doctrine: data-structure-first, kill over-engineering, demand proof
- 633 Works with setup
Routes code exploration to a code-graph MCP server instead of grep/read.
- 634 Tested · Works
Enforces Vue 3.4+ TypeScript patterns: defineModel, typed emits, Testing Library over test-utils.
- 635 Works with setup
Workbench Agent Install Unifier
Safety contract for syncing skills/MCPs/AGENTS.md across coding-agent CLIs via agent-install.
- 636 Works with setup
Documents Frank's personal ACOS system — not a generic Claude Code skill-authoring guide.
- 637 Works with setup
QEMU/GDB/Frida runbook for tracing and hooking a binary's actual runtime behavior.
- 638 Tested · Works
Tool-first codebase health scan: dead code, test quality, complexity, security, architecture
- 639 Works with setup
JSON-RPC bridge that lets Claude Code delegate tasks and PR reviews to Codex and read the results back.
- 640 Works with setup
Local auth-split proxy that gives OpenAI Codex App priority (Fast) tier on third
- 641 Works with setup
Runs Claude, Gemini and GPT in parallel and synthesizes a debated consensus answer.
- 642 Tested · Works
Scaffolds Cursor .mdc rule files in .cursor/rules/ with correct frontmatter and scope.
- 643 Works with setup
Playwright-backed browser automation with a persistent page server, but its own scripts live outside this repo.
- 644 Works with setup
Discover a codebase's conventions and scaffold rules, skills, wiki and CI gates for AI coders
- 645 Works with setup
Explains a Rails file or a whole user flow, starting from the git history
- 646 Works with setup
Opinionated Expo React Native scaffold: RevenueCat paywall, AdMob, i18n, NativeTabs
- 647 Tested · Works
Architecture prose that justifies every call and defines a check per slice
- 648 Works with setup
Cross-model adversarial PR review via Codex/Copilot/Cursor/RepoPrompt inside flow-next.
- 649 Works with setup
Prompt recipes and antipatterns for delegating coding work to Gemini 2.5
- 650 Works with setup
GPIO pin assignment, conflict validation, and init code for Pi and ESP32.
- 651 Works with setup
8-phase feature/epic execution checklist wrapped around an external ruflo swarm MCP.
- 652 Tested · Works
Converts HTML+CSS into Elementor JSON, with a verifier that checks the cascade
- 653 Tested · Works
Hugging Face Inference patterns for TypeScript: streaming, embeddings, errors
- 654 Works with setup
Routes all KiCAD schematic/PCB edits through the Konnect MCP server instead of raw text edits
- 655 Works with setup
Lift any runbook, SKILL.md or chat log into a runnable Flowtrace DAG and drive the full run lifecycle
- 656 Works with setup
Manage the Postman API lifecycle: collections, specs, tests, monitors, mocks
- 657 Works with setup
Generates and validates installable Hermes Agent profile repos
- 658 Works with setup
Security checklist for safe commands, URLs, secrets and supply chain
- 659 Works with setup
Hybrid semantic + BM25 code search that intercepts and redirects grep
- 660 Works with setup
Routes agent credential ops through the Symaira Vault MCP server instead of shell
- 661 Works with setup
Syncs and PII-scrubs skills from a private canonical repo into a public one -- author-only.
- 662 Works with setup
4-agent harness for translating a codebase to a new language/substrate line by line.
- 663 Tested · Works
Rewrites docker-compose for xCloud: no build:, no port 80/443, one exposed port
- 664 Works with setup
Severity-triaged git-diff review across security/perf/pattern/architecture lenses.
- 665 Works with setup
Operator guide for the botfile CLI, but the skill file ships buried in an examples/tutorial fixture, not a top-level skills folder
- 666 Works with setup
Claude Code plugin: 39 guided skills plus an auto-configured MCP for the Butterbase backend-as-a-service
- 667 Works with setup
Condensed cheat-sheet on Claude Code subagents, skills, hooks, and MCP for improving your own workflow.
- 668 Works with setup
Auto-loops Codex to review Claude's plans and code via Byterover shared memory
- 669 Works with setup
Mechanically generates an Agent Skill markdown file from any CLI's --help output or an MCP server's tool list.
- 670 Works with setup
Wraps a third-party stealth-Chromium tool for anti-bot scraping, but needs a 206 MB proprietary binary the repo doesn't ship.
- 671 Works with setup
Stack-neutral pre-commit review against SOLID, DRY, security and dead-code
- 672 Works with setup
Executes a feature's implementation plan task-by-task and checks the result against its design doc.
- 673 Works with setup
Scaffolds a feature.md + branch + tracker entry for the Disciplined Agentic Engineering pipeline.
- 674 Works with setup
Coordinates a Carmack-style AI plan review for Flow Next specs via Codex, Copilot, Cursor, or RepoPrompt.
- 675 Works with setup
Enforces BLoC layering, Equatable states and design-system constants in Flutter
- 676 Works with setup
Phased Nova-to-Filament v5 migration playbook with a 25-row field mapping table
- 677 Works with setup
Sets up, runs, and debugs the AutoResearchClaw 23-stage autonomous paper-writing CLI pipeline.
- 678 Works with setup
Umbrella for sc-* deploy skills: Dokploy, Convex, Vercel, GitHub, Hostinger DNS
- 679 Works with setup
Terminal control of the user's real, authenticated Firefox - but the skill folder ships zero setup help.
- 680 Works with setup
Slash-command orchestrator for a research-plan-impl-review dev cycle with fix loop
- 681 Works with setup
Ships a solid embedded quality checklist, but the two reference files holding the actual '245 rules' don't exist in the repo.
- 682 Works with setup
Netresearch's own standard for skill-repo layout, split licensing, and release CI — but its validator script crashes on macOS.
- 683 Works with setup
On-device Apple Silicon TTS CLI with voice cloning, auto-chunking and resume for long documents.
- 684 Works with setup
Turns audit findings into atomic commits, one test run per fix
- 685 In test queue
A 29-line review-prompt template living in skill-up's e2e test fixtures
- 686 Works with setup
Turns a one-line requirement into a TRD via seven architecture decision lenses
- 687 Works with setup
Advisory Mining (find-cve-agent)
Multi-agent CVE-hunting plugin with OSV/NVD lookups and false-positive gates
- 688 Tested · Didn't pass
ACF Custom Fields REST Exposer
WordPress ACF-to-REST note that ships no code, just truncated scraped prose
- 689 Tested · Didn't pass
Git hooks and scanners that mechanically enforce agent coding rules
- 690 Tested · Didn't pass
Alias for the Alfred Dev plugin; its installer widens your permissions
- 691 Tested · Didn't pass
Thin API wrapper for Ardent Postgres branching, gated behind a proprietary vm0.ai connector
- 692 Tested · Didn't pass
Autonomous SPARC build contract: phase gates, self-grade >=95 loop, checkpoints
- 693 Tested · Didn't pass
Pinned Context7 library IDs for fetching current Cesium and 3D Tiles docs
- 694 Tested · Didn't pass
Meta-orchestrator that points at other coding skills it does not ship
- 695 Tested · Didn't pass
Converts agent definitions between Claude, OpenAI, CrewAI, and other frameworks.
- 696 Tested · Didn't pass
Hands large cross-file analysis to the Gemini CLI via a shared bridge script
- 697 Tested · Didn't pass
Bumps a plugin's semver in marketplace.json via jq, with major/minor/patch inference rules.
- 698 Tested · Didn't pass
Routing table that dispatches OpenSkills dev tasks to its own project skills and subagents
- 699 Tested · Didn't pass
Kubernetes/GitOps/Terraform/CI-CD router whose 59 reference playbooks live outside the installable skill folder.
- 700 Tested · Didn't pass
Steers Claude to PyWry's MCP tools for Python UIs instead of Flask/Streamlit/Electron
- 701 Tested · Didn't pass
Reverse Engineer Anything (REA)
Routing layer for the REA reverse-engineering toolchain; needs its own CLI
- 702 Tested · Didn't pass
RubyCritic Code Quality Analysis
Wraps the RubyCritic gem to grade Ruby/Rails files A-F and prioritize refactoring by churn.
- 703 Tested · Didn't pass
MDX lint and semantic-commit git publishing — broken in practice
- 704 Tested · Didn't pass
Batch-grades every SKILL.md in a repo, worst-first, via skill-forge-judge.
- 705 Tested · Didn't pass
SvelteKit 2 + Svelte 5 + Tailwind v4 Integration
Bundled doc-search reference for SvelteKit 2, Svelte 5, and Tailwind v4
- 706 Tested · Didn't pass
Linux triage cheat sheet: load, memory, disk, systemd services, top offenders
- 707 Tested · Didn't pass
Opinionated multi-agent build workflow: spec, parallel research, concurrent revi
- 708 Tested · Didn't pass
TanStack AI cheat sheet whose core API does not match the published npm packages
- 709 Tested · Didn't pass
Orchestrates a scientific-method debug session in an isolated subagent
- 710 Tested · Didn't pass
Governed runtime that freezes requirements and gates execution behind approvals
- 711 In test queue
Orchestrates multi-phase, resumable cc-fleet subagent runs from a JS script.
Can't find what you need?
Request a skill — we'll test or build it
Tell us the job. Within a few days you get back a link to a tested skill that already does it, a SKILL.md you can build from, or a ready-to-use prompt. Popular asks get built for the catalog.
Request received — check your inbox for confirmation.
Describe the need in a sentence or two, and use a real email.