TokenTracker
xiufengsun/TokenTracker/CLAUDE.md
src/lib/rollout.js` — the source of truth, not this file. ## Frequently used commands ```bash npm test # node --test test/*.test.js (209 files) node --test test/ .test.js # single test file
How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
xiufengsun/TokenTracker/CLAUDE.md
src/lib/rollout.js` — the source of truth, not this file. ## Frequently used commands ```bash npm test # node --test test/*.test.js (209 files) node --test test/ .test.js # single test file
happier-dev/happier/.claude/CLAUDE.md
corridor, a full QA matrix, a full review) end to end: analysis, implementation, tests, validation, ledger updates. Micro-slicing (one-boolean extractions) provably grows god-files instead of shrinking them
Jesseovo/last30days-skill-cn/CLAUDE.md
skills/last30days/`: generated installable Agent Skill payload. - `scripts/lib/render.py`: Markdown, JSON context, and HTML report rendering. - `tests/`: focused regression tests. Only edit the root `SKILL.md` and `scripts/` tree. Regenerate the installable payload
microsoft/rnx-kit/incubator/fork-sync/CLAUDE.md
assisted conflict resolution. ## Quick Reference ```bash yarn build # Compile TypeScript to lib/ yarn test # Run all tests (fast: skips slow integration tests) yarn test:all # Run all tests including slow
feigeCode/navop/CLAUDE.md
Commands ```bash # Run the application cargo run -p main # Build cargo build # Run all tests cargo test --all # Run tests for a specific crate cargo test -p gpui-component cargo
relaticle/relaticle/CLAUDE.md
Adding a writable field or a relation include to one of them fails that test until the others follow. A query predicate over one model's columns has one owner
BayramAnnakov/claude-reflect/CLAUDE.md
Skill definitions for /reflect, /reflect-skills, /skip-reflect, /view-queue SKILL.md → Context provided when plugin is invoked tests/ → Test suite (pytest) ``` ### Data Flow 1. User prompt → `capture_learning.py` (UserPromptSubmit hook) → `~/.claude/learnings-queue.json` 2. `/reflect` command
lahfir/agent-desktop/CLAUDE.md
Commands ```bash cargo build # Debug build cargo build --release # Release build (<15MB target) cargo test --workspace # Run every test (lib, binary, and integration targets) cargo test --lib --workspace # Library unit
lst97/claude-code-sub-agents/CLAUDE.md
Stage N: [Name] **Goal**: [Specific deliverable across the stack] **Success Criteria**: [User story + passing tests] **Tests**: [Unit, integration, E2E coverage] **Status**: [Not Started|In Progress|Complete] ``` - Update status after each
NVIDIA/trt-samples-for-hackathon-cn/CLAUDE.md
print commands only python3 tests/run_tests.py --tags plugin # filter by tag ``` **Run the full test suite (unified runner):** ```bash python3 tests/run_tests.py ``` **Run pytest tests (NetworkSerialization):** ```bash pytest tests/NetworkSerialization/ pytest tests/NetworkSerialization/test_convolution.py # single
manzaltu/claude-code-ide.el/CLAUDE.md
imenu - `claude-code-ide-diagnostics.el` - Flycheck integration - `claude-code-ide-transient.el` - Transient menu interface - `claude-code-ide-debug.el` - Debug logging utilities - `claude-code-ide-tests.el` - ERT test suite with mocks ## Hooks This project uses Claude Code hooks to automatically maintain code quality
aws/aws-lambda-dotnet/Libraries/src/Amazon.Lambda.DurableExecution/CLAUDE.md
listed in `README.md` for the shared model — this SDK deliberately mirrors their semantics. ## Build & test Targets `net8.0;net10.0` (`DefaultPackageTargets` in `buildtools/common.props`). `TreatWarningsAsErrors` is on everywhere, and the main library
JetBrains/MPS/CLAUDE.md
Developers typically use two environments against the same checkout: - IntelliJ IDEA for editing, debugging, testing, and inspecting Java/Kotlin code, including code generated by MPS. - JetBrains MPS for editing languages, models
ghostwright/ghost-os/CLAUDE.md
stderr). Never use print (corrupts MCP stdout). - No force unwraps except in tests. - Error responses include a `suggestion` field telling the agent what to do next. - No app-specific hacks
patrickchugh/terravision/docs/CLAUDE.md
output-dir ./diagrams # Debug mode (exports tfdata.json) poetry run terravision draw --source --debug ``` ### Testing ```bash # Run all tests poetry run pytest tests -v # Run non-slow tests (for pre-commit
ddalcu/mlx-serve/app/CLAUDE.md
pure `QuickLauncherLogic` | | `Services/{ImagePreprocessor,BrowserManager,ServerManager,TestServer,AgentMemory}.swift` | Image prep, WKWebView, server lifecycle, test server :8090, agent context | | `Services/{LaTeXSegmenter,LaTeXRenderer}.swift` + `Views/ChatView.swift` (`MarkdownText`) | Assistant-response TeX: streaming-safe
ddalcu/mlx-serve/CLAUDE.md
models-media,app}.md` — full war stories behind every rule in `## Rules`. - `tests/CLAUDE.md` — integration-test matrix. `app/CLAUDE.md` — Swift app layout + rules (auto-load in their dirs). - Skills: `/release` (pre-release
ddalcu/mlx-serve/tests/CLAUDE.md
tests/ — test matrix & guard reference (auto-loaded when working here) Root CLAUDE.md carries the TDD policy; this file is the full matrix of integration/guard scripts and what each pins. ## Testing
Asymptote-Labs/agent-beacon/CLAUDE.md
dist-firefox/`), and every extension API call goes through the `src/shared/browser.ts` shim (a unit test fails on direct `chrome.*` use). The Firefox Gecko ID `browser-collector@agent-beacon.asymptotelabs.ai` is pinned and must never
e2b-dev/infra/CLAUDE.md
staging # Setup local development stack (PostgreSQL, Redis, ClickHouse, monitoring) make local-infra ``` ### Building & Testing ```bash # Run all unit tests across packages make test # Run integration tests make test-integration # Build
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.