What is clawstash AGENTS.md?

Project instructions for coding agents working on fo0/clawstash. They require reading the project’s main guide first and limit GitNexus, a code-analysis tool, to read-only use.

Use them when navigating, analyzing, or changing the fo0/clawstash repository, especially when GitNexus is involved.

They reduce the risk of ignoring project conventions or accidentally changing files during code analysis. They also make the project’s safety rules explicit.

clawstash AGENTS.md — Instructions file by fo0

Install

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.

agentmods

npx agentmods add instructions/fo0/clawstash/agents-md

Clone the repo

git clone --depth 1 https://github.com/fo0/clawstash

Made for: Codex, OpenCode.

Per session 509 This file is loaded in full into every session.

When invoked 509 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What 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	$0.00509	$0.00509
Opus 5	$0.00254	$0.00254
Sonnet 5	$0.00102	$0.00102
Haiku 4.5	$0.00051	$0.00051

Measured 2d ago against content hash 0cd9416af6d2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

clawstash AGENTS.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.

AGENTS.md · 31 lines

What it actually says

AGENTS.md — Agent Guide

Primary guide: read CLAUDE.md first. It is the single source of truth for project structure, commands, conventions, workflows, and skills. This file carries the canonical, verbatim text of the non-negotiable GitNexus guardrail — CLAUDE.md keeps a condensed version and agent_docs/gitnexus.md only points here.

GitNexus — Read-Only Analysis Policy (non-negotiable)

GitNexus is an analysis/read-only tool. It must never write to the repository.

Allowed: read-only MCP tools only — gitnexus_query, gitnexus_impact, gitnexus_context, gitnexus_detect_changes, and status/list. Use these to understand code, assess blast radius, and navigate. They never modify files.
Forbidden: creating, scaffolding, regenerating, or editing ANY file as a side effect of GitNexus — in particular .claude/skills/** (including GitNexus's own gitnexus/* skill files), CLAUDE.md, AGENTS.md, docs/wiki/**, or anything else.
gitnexus analyze / index: only run when the index is genuinely missing or stale AND it is required for the current task. When you do, pass --skip-agents-md and treat it as index-only: it must NOT touch tracked files. If it modifies .claude/skills/**, CLAUDE.md, AGENTS.md, or any other tracked file, revert those changes immediately (git checkout -- <paths>). The index itself (.gitnexus/) stays gitignored and uncommitted.
Never include GitNexus-generated skill/doc edits in a commit or PR. They are out of scope for every task unless I explicitly ask for them.
Pre-commit guard: before any commit, run git status and verify no unexpected .claude/**, CLAUDE.md, AGENTS.md, or agent-doc changes are staged. If there are and they weren't the point of the task, revert them and proceed.

Changes

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 · 31 lines · 509 tokens per session scan A 0cd9416af6d2

Subscribe to this mod's changes

clawstash AGENTS.md is an instructions file published in the GitHub repository fo0/clawstash (1 stars, last pushed yesterday), licensed MIT. It adds 509 tokens to every session, about $0.0025 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.

ai-tools api developer-tools mcp mcp-server model-context-protocol

clawstash AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

AGENTS.md — Agent Guide

GitNexus — Read-Only Analysis Policy (non-negotiable)

What this file has done since we first saw it

Other instructions, from other repositories

pricing-api AGENTS.md

Arch-AI-Tools AGENTS.md

obsidian-mcp-server AGENTS.md

Agent-Span AGENTS.md

obsidian-pkm-plugin CLAUDE.md

pdf-toolkit-mcp CLAUDE.md