Pack · v0.1.0
AI / agent engineer (LLM apps, evals, MCP)
AI coding-agent instructions for: Building LLM-powered features and agents in TypeScript: a swappable model interface with deterministic fakes, budgeted agent loops, human…
About this pack
AI coding-agent instructions for: Building LLM-powered features and agents in TypeScript: a swappable model interface with deterministic fakes, budgeted agent loops, human approval for risky tools, eval harnesses with CI gates, and MCP servers with safe design.
What is inside (21 files)
- .cursor/rules/00-core.mdc
- .cursor/rules/10-agent-code.mdc
- AGENTS.md
- CLAUDE.md
- CONTRIBUTING.md
- LICENSE
- README.md
- docs/SOURCES.md
- docs/STARTER_CHECKLIST.md
- skills/agent-loop-guardrails/SKILL.md
- skills/eval-harness/SKILL.md
- skills/llm-client-and-fake-tests/SKILL.md
- skills/tool-and-mcp-server-design/SKILL.md
- templates/src/agent.test.ts
- templates/src/agent.ts
- templates/src/evals.test.ts
- templates/src/evals.ts
- templates/src/llm.ts
- templates/src/mcp-server.ts
- templates/src/mcp.test.ts
- templates/tsconfig.json
Good to know
- How do I get it?
- Sign in, pay once with Stripe Checkout, then download the zip from My library. Links are short-lived; you can re-download any time.
- Does it work with my agent?
- Packs use plain AGENTS.md, CLAUDE.md, Cursor rules and SKILL.md files, so any agent that reads those formats can use them.
- Are claims verified?
- Unverified items are labelled [UNVERIFIED] inside each pack. Not affiliated with any tool or engine vendor.