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/finos/git-proxy/claude-mdgit clone --depth 1 https://github.com/finos/git-proxyWhat 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 | $0.00153 | $0.00153 |
| Opus 5 | $0.00077 | $0.00077 |
| Sonnet 5 | $0.00031 | $0.00031 |
| Haiku 4.5 | $0.00015 | $0.00015 |
Grade A, and why
git-proxy 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 yesterday.
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.
What it actually says
Claude Instructions
Read and follow AGENTS.md first. It is the canonical project guide for this repository.
Additional Claude-specific notes:
- Apply
AGENTS.mdas the source of truth. If this file andAGENTS.mdever differ,AGENTS.mdwins. - Keep changes aligned with the chain-of-processors architecture. Do not bypass approval gating,
blockForAuth, waiting-authorization checks, or audit logging. - Prefer the existing build, test, lint, and format commands listed in
AGENTS.md. - Project slash commands are defined as skills in
.agents/skills/. - Treat
.claude/worktrees/as local machine state. Do not rely on it or commit changes from it.
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.
- yesterday First seen · 12 lines · 153 tokens per session scan A 025a7c7e636a
git-proxy CLAUDE.md is an instructions file published in the GitHub repository finos/git-proxy (243 stars, last pushed 2d ago), licensed Apache-2.0. It adds 153 tokens to every session, about $0.0008 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 2026-08-30.
Other instructions, from other repositories
agent-skills AGENTS.md
Instructions for fluxcd/agent-skills, covering agent instructions, repository layout, working on existing skills, running skill evals and adding a new skill.
bex CLAUDE.md
Instructions for bex-co/bex, covering claude.md, repo map, commands, local cluster workflow and environment variables.
radar CLAUDE.md
Instructions for skyhook-io/radar, covering claude.md, project overview, code comments, release and publishing authorization and reference docs — must read before making changes.
kafka-backup CLAUDE.md
Instructions for osodevops/kafka-backup, covering project overview, key problems solved, repository structure, build & test commands and build the project.
agent-skills CLAUDE.md
Instructions for fluxcd/agent-skills: Load the skill-creator for tasks in this repo.
kanea AGENTS.md
Instructions for m18h/kanea, covering agents.md: kanea, what this is, current status, recurring design rules and binding constraints (never violate these).