Is the Implement Factory skill free?

Yes. The skill itself is free from rsmdt/the-startup. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Implement Factory SkillProof Score?

7.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 4/5.

How do I install Implement Factory?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Implement Factory with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Implement Factory

Orchestrator spawning isolated code and eval subagents to build spec manifests via TDD

by rsmdt · rsmdt/the-startup

Works with setup

Install — copy the command Download source ↗

Test report

Verdict: Works with setup
Score: 7.6/10
Tested: Jul 31, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · 19eee89

Located SKILL.md at plugins/start/skills/implement-factory/; frontmatter parses with name+description (plus user-invocable:false, argument-hint). All three referenced files (reference/code-agent.md, eval-agent.md, output-format.md) returned HTTP 200, and the named dependency skills specify-meta and validate both exist as siblings in the plugin. No security smells found. Could NOT run the factory loop: it requires a scaffolded .start/specs/NNN/ manifest, the specify-meta skill to resolve it, genuine isolated subagent spawning with enforced information barriers, and a live service with a /health endpoint — none reproducible in a single evaluation context, so no output artifact was produced. Verified the repo's own eval fixtures (eval-1-happy-path manifest.md, code-agent template) which confirm the documented structure is real.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 5/10
Docs & honesty 4/5

What Implement Factory does

A factory-loop orchestrator that implements a multi-unit spec manifest by spawning isolated code-agent and evaluation-agent subagents, enforcing an information barrier so code agents never see test scenarios and eval agents never see source. It runs a TDD implement-then-evaluate loop against a live service, retrying units below a satisfaction threshold. Triggers on high-complexity, multi-component implementation work driven by a .start/specs manifest with parallel execution groups and holdout-scenario evaluation.

How to install Implement Factory

git clone --depth 1 https://github.com/rsmdt/the-startup.git /tmp/implement-factory-src
mkdir -p ~/.claude/skills
cp -R /tmp/implement-factory-src/plugins/start/skills/implement-factory ~/.claude/skills/implement-factory
# NOTE: frontmatter sets user-invocable:false — designed to be spawned by the /start plugin's
#   implement router, not invoked directly by the user.
# DEPENDS ON the surrounding ecosystem: sibling skills specify-meta + validate, and a scaffolded
#   .start/specs/NNN-name/ manifest (manifest.md + units/ + scenarios/). Installing this skill
#   alone is not enough to run it.
# For real use, install the whole plugin/repo (https://github.com/rsmdt/the-startup), not just this dir.

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Implement Factory

/implement-factory Orchestrator spawning isolated code and eval subagents to build spec manifests via TDD

It also activates on plain-language prompts like these:

Run the factory loop on this implementation manifest
Parallelize these workstreams for the feature build
Evaluate holdout scenarios across these components

Frequently asked questions

Is the Implement Factory skill free?: Yes. The skill itself is free from rsmdt/the-startup. SkillProof publishes the install command and an independent test verdict at no cost.
Does Implement Factory work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 31, 2026. Verdict: Works with setup. Located SKILL.md at plugins/start/skills/implement-factory/; frontmatter parses with name+description (plus user-invocable:false, argument-hint). All three referenced files (reference/code-agent.md, eval-agent.md, output-format.md) returned HTTP 200, and the named dependency skills specify-meta and validate both exist as siblings in the plugin. No security smells found. Could NOT run the factory loop: it requires a scaffolded .start/specs/NNN/ manifest, the specify-meta skill to resolve it, genuine isolated subagent spawning with enforced information barriers, and a live service with a /health endpoint — none reproducible in a single evaluation context, so no output artifact was produced. Verified the repo's own eval fixtures (eval-1-happy-path manifest.md, code-agent template) which confirm the documented structure is real.
What is the Implement Factory SkillProof Score?: 7.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 4/5.
How do I install Implement Factory?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Implement Factory with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.