agentleFS
Sign inStart free
furkantanyolfurkantanyol/hevy-coach/CLAUDE.md

hevy-coach

CLAUDE.md2 starsChanged 2 months agoMITView on GitHub

Two npm packages built on the Hevy public API: The coaching persona and setup guides for Claude / ChatGPT / Gemini live in prompts/. This file is for working on the code. Build before typecheck: the server resolves the client through exports → dist/. pnpm validate does this in the right order.

  • Reads credentials
  • Installs packages
  • In this repositoryOne other file it gives its agents.

The file

Create a free account to mount
Files
CLAUDE.md

hevy-coach monorepo

Two npm packages built on the Hevy public API:

  • packages/hevy-sdk → hevy-sdk: typed fetch-only client (Node + React Native).
  • packages/mcp-server → hevy-coach: MCP server with coaching tools, runs with npx hevy-coach.

The coaching persona and setup guides for Claude / ChatGPT / Gemini live in prompts/. This file is for working on the code.

Commands

pnpm install
pnpm validate          # biome format+lint, build, typecheck, tests. The quality gate.
pnpm generate:types    # regenerate packages/hevy-sdk/src/generated/schema.ts from docs/api/*.json
pnpm --filter hevy-coach dev   # run the stdio server from source (needs .env with HEVY_API_KEY)
pnpm --filter hevy-coach pack:mcpb   # Claude Desktop extension → packages/mcp-server/hevy-coach.mcpb

Build before typecheck: the server resolves the client through exports → dist/. pnpm validate does this in the right order.

Rules

  • TypeScript 5.9 pinned (ADR 0002). Strict, no any.
  • Live API beats the spec. Spec bugs are patched in packages/hevy-sdk/scripts/generate-types.mjs, each with a comment citing docs/api/probes-<date>.md. Never edit the snapshot JSON or src/generated/.
  • The client has zero runtime dependencies and no Node-only APIs. fetch only.
  • Tests mock fetch via options.fetch; never hit the live API in tests.
  • Writes against the live API only to routines named [TEST] ....
  • HEVY_API_KEY lives in a gitignored .env at the repo root. Never commit it.
  • Decisions with a "why" go in docs/adr/. Keep them short.
  • Small conventional commits.

Layout

docs/api/        OpenAPI snapshots + probe results (the evidence)
docs/adr/        decisions
prompts/         COACH.md persona, CHATGPT.md, GEMINI.md, training-history.md template
packages/hevy-sdk/src   client.ts http.ts pagination.ts sync.ts errors.ts types.ts generated/
packages/mcp-server/src    cli.ts http.ts index.ts tools/ utils/

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free