Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/hanselhansel/context-cli/claude-mdgit clone --depth 1 https://github.com/hanselhansel/context-cliWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.02591 | $0.02591 |
| Opus 5 | $0.01295 | $0.01295 |
| Sonnet 5 | $0.00518 | $0.00518 |
| Haiku 4.5 | $0.00259 | $0.00259 |
Grade A, and why
context-cli CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context CLI v2.0.0 — CRITICAL CONTEXT
Scoring: Content=40 | Robots=25 | Schema=25 | llms.txt=10 (total=100)
Bots: GPTBot, ChatGPT-User, Google-Extended, ClaudeBot, PerplexityBot, Amazonbot, OAI-SearchBot + 6 more
Commands: pytest | ruff check src/ tests/ | mypy src/ | make ci
CURRENT PHASE: 2.0.0 (Rebrand — pivot from Context CLI to Context CLI)
AGENT TEAMS: MANDATORY for ALL phases. No exceptions. See "Agent Teams" section below.
Session Workflow (MANDATORY — enforced by hooks)
- TEST — Write comprehensive tests FIRST (TDD, 100% coverage)
- CODE — Implement minimum to pass tests
- REFACTOR — Clean up: split large files (>300 LOC), extract modules, simplify
- LINT — Auto-lint via PostToolUse hook on every file change
- VERIFY —
pytest --cov=context_cli --cov-fail-under=100(100% mandatory) - COMMIT —
git add+git commit+git push origin main(every green feature) 6.5. README — If releasing a version: update README Features, CLI Usage, and Bots sections - VERSION — Patch bump + tag + push when feature is significant
- CONTEXT — Check /context; if >70% →
/compact8.5. AGENTS — If session resumes with active worktrees: check agent branches for commits, merge completed work, clean up worktrees
Versioning
Minor bump per phase: 0.3.0(A1) → 0.4.0(A2) → 0.5.0(A3) → 0.6.0(A4) → 0.7.0(B0) → 0.8.0(B1) → 0.9.0(B2) → 0.10.0(B3) → 0.11.0(B4) → 0.12.0(B5) → 1.0.0 → 2.0.0(Rebrand)
Patch within phases (2.0.1, 2.0.2, ...). Every v* tag auto-publishes to PyPI.
--- END CRITICAL HEADER (above MUST survive compaction) ---
Tech Stack
- Language: Python 3.10+
- CLI: Typer + Rich
- HTTP: httpx (async) with retry logic
- HTML parsing: BeautifulSoup4
- Crawling: crawl4ai (headless browser)
- Data models: Pydantic v2
- MCP server: FastMCP
- Testing: pytest + pytest-asyncio + pytest-cov
- Linting: Ruff (line-length=100)
- Type checking: mypy (with Pydantic plugin)
- CI: GitHub Actions (test + lint + mypy, Python 3.10/3.11/3.12 matrix)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 185 lines · 2,591 tokens per session scan A bfd3b0fffad3
context-cli CLAUDE.md is an instructions file published in the GitHub repository hanselhansel/context-cli (4 stars, last pushed 5mo ago), licensed MIT. It adds 2,591 tokens to every session, about $0.0130 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
specification.website CLAUDE.md
Claude Code instructions for jdevalk/specification.website, covering what this project is, status discipline, changelog, cardinal rules for content and architecture quick reference.
oxc copilot-instructions.md
Instructions for oxc-project/oxc: When reviewing pull requests in this repository.
geo-optimizer-skill AGENTS.md
Instructions for Auriti-Labs/geo-optimizer-skill, covering agents.md, project context, repository map, architectural rules and security rules.
technocore-chat AGENTS.md
Instructions for flop-labs/technocore-chat: CI runs exactly these — run them before pushing.
seo-graph AGENTS.md
Instructions for jdevalk/seo-graph, covering agents.md — seo-graph, what this library does, architecture, installation and astro projects — install both.
blz AGENTS.md
Instructions for outfitter-dev/blz, covering blz repository instructions for ai agents, important, working memory, use blz and 🚀 quick start for agents.