formatting and notebook structure.
## Slash Commands
These commands are available in Claude Code and CI:
- `/notebook-review` - Review notebook quality
- `/model-check` - Validate Claude model references
- `/link-review` - Check links in changed files
patches.
## GitHub
### Pull Requests
Use the `gh-create-pr` skill. Fallback: read `.agents/skills/gh-create-pr/SKILL.md` directly.
### CodeReview
When reviewing a GitHub PR, do NOT run `pnpm lint` / `pnpm test` / `pnpm format
Academic Research Skills
A suite of Claude Code skills for rigorous academic research, paper writing, peer review, and pipeline orchestration.
## Skills Overview
| Skill | Purpose | Key Modes |
|-------|---------|-----------|
| `deep-research` v2.12.1
AGENTS.md` as loaded. Claude Code does not reliably inline that import.
Before any planning, coding, reviewing, or answering a project question, you MUST call the Read tool on the repo
state the change actually carries. Describe the code
as it stands: do not narrate what earlier commits tried or what a review round changed. Naming the
merged pull request that
Important:** When mentioning agentic coding or AI tooling in CVs/cover letters, explicitly reference **Claude Code** by name.
## Verification Checklist
After creating or updating a CV or cover letter, re-read
code (ruff check, mypy)
./scripts/validate.sh
```
Both scripts must pass with no errors before codereview.
**PR Title Format:**
PR titles must follow one of these formats:
- `type: description` — e.g., `feat
push
```
> The parent repo stores a commit SHA pointer to each submodule, not the code itself. Always push from inside the submodule first, then update the pointer in the parent
later in a separate lane.
Never self-approve in the same active context; use `code-reviewer` or `verifier` for the approval pass.
Before concluding: zero pending tasks, tests passing, verifier
setting read-only-mode
# Interactive mode
npx claude-code-templates@latest
```
### Component Development
#### Adding New Components
**CRITICAL: Use the component-reviewer agent for ALL component changes**
When adding or modifying
CLAUDE.md - Project Context for Claude Code
## Project Overview
code-review-graph is a local-first knowledge graph for codereview. It parses a repository with Tree-sitter (plus targeted fallbacks
trusting forwarding headers only from a loopback reverse proxy. Preserve this when touching request/IP/rate-limit code.
- Security-sensitive env: `JWT_SECRET` (session cookie), `INITIAL_PASSWORD` (default `123456` — must override
scripts/cmux-debug-cli.sh`, never `/tmp/cmux-cli`.
Never report raw `.app` paths or `file://` URLs.
- App-linked code (`Sources/`, `CLI/`, `TunnelExtension/` and their packages) must
remain [Swift 6.0 compatible](skills/cmux-architecture/references/swift-6-0-compatibility.md).
## Area instructions
Read these
deviations from upstream (6 cosmetic, 17 safety/hardening)**, found across ten rounds of adversarial codereview rather than assumed safe from the surface docs. The **numbered list in `engineering/skillopt-sleep/README.md`'s "Deviations
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
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.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.