Convex Agent
Teaches the exact @convex-dev/agent API for threads, streaming, and tool-calling agents.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 15, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · c055c02
Real repo-vs-doc bug found: SKILL.md links references/RAG.md but the file on disk is actually RAG.mD, which 404s on a raw GitHub fetch (case-sensitive) even though 10 other reference links resolve fine. On the actual task (wire up a Convex support-chat action), the skill output used the real @convex-dev/agent package (saveMessage + asTextAction, persistent threads) where the baseline reinvented a hand-rolled messages table and manual OpenAI call with no streaming or retry-safe message IDs — a substantive, non-cosmetic difference this skill exists to prevent.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 4/5
- Triggers reliably 5/5
- Output vs. baseline 9/10
- Docs & honesty 4/5
What Convex Agent does
Reference skill for the Convex Agent component: persistent threads, message storage, streaming, tool calling, and durable multi-step agent workflows on the Convex backend. Triggers when a user is building a chat interface, AI assistant, RAG feature, or any LLM-powered feature inside a Convex project; not relevant outside Convex or for non-Convex auth/backend questions.
How to install Convex Agent
git clone https://github.com/PolarCoding85/convex-agent-skillz.git
cd convex-agent-skillz
mkdir -p ~/.claude/skills
cp -r .claude/skills/convex-agent-skill ~/.claude/skills/convex-agent
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Convex Agent
-
/convex-agentTeaches the exact @convex-dev/agent API for threads, streaming, and tool-calling agents.
It also activates on plain-language prompts like these:
-
Wire up a Convex mutation that persists chat messages to a support agent -
Add streaming responses to my Convex AI assistant using the agent component -
Build a RAG feature in my Convex backend with persistent threads
Frequently asked questions
- Is the Convex Agent skill free?
- Yes. The skill itself is free from PolarCoding85/convex-agent-skillz. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Convex Agent work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 15, 2026. Verdict: Tested · Works. Real repo-vs-doc bug found: SKILL.md links references/RAG.md but the file on disk is actually RAG.mD, which 404s on a raw GitHub fetch (case-sensitive) even though 10 other reference links resolve fine. On the actual task (wire up a Convex support-chat action), the skill output used the real @convex-dev/agent package (saveMessage + asTextAction, persistent threads) where the baseline reinvented a hand-rolled messages table and manual OpenAI call with no streaming or retry-safe message IDs — a substantive, non-cosmetic difference this skill exists to prevent.
- What is the Convex Agent SkillProof Score?
- 8.8/10 — installs cleanly 4/5, triggers reliably 5/5, output vs. baseline 9/10, docs & honesty 4/5.
- How do I install Convex Agent?
- 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 Convex Agent 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.