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/stanmaygo/dotsweep/claude-mdgit clone --depth 1 https://github.com/stanmaygo/dotsweepWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/stanmaygo/dotsweep/claude-md)<a href="https://agentmods.dev/instructions/stanmaygo/dotsweep/claude-md"><img src="https://agentmods.dev/badge/instructions/stanmaygo/dotsweep/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00640 | $0.00640 |
| Opus 5 | $0.00320 | $0.00320 |
| Sonnet 5 | $0.00128 | $0.00128 |
| Haiku 4.5 | $0.00064 | $0.00064 |
Grade A, and why
dotsweep 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 4d 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.
This is a copy
100% identical to dotsweep AGENTS.md — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dotsweep-skill
The published agent skill for dotsweep, and its packaging. Nothing else lives here.
AGENTS.md is a byte-identical copy of this file: Codex, Cursor and Warp read
that name, Claude Code reads this one, and the validator holds them equal so
they are one document rather than two.
This repo is a mirror, not the source
skills/dotsweep/SKILL.md is generated by copying, from the engine
repository that also holds the Go CLI and the Cloudflare Worker. Editing it
here is how the two come to disagree: the next sync overwrites the change, and
the hosted dotsweep.com/skill.md — which is what the paste-one-message
install actually fetches — keeps serving the other version. Two installs of the
same skill then behave differently with nothing reporting it.
Edit the prompt in the engine repo, then run scripts/sync-skill.sh here.
What must stay true
-
SKILL.mdat the root andskills/dotsweep/SKILL.mdhold the same bytes.npx skills addreads the root path and a manual installer reads the directory, so both have to be the prompt.This was a symlink, and a symlink does not survive publication:
raw.githubusercontent.comserves a link's target as the file body, so the published rootSKILL.mdreturned the 24-byte stringskills/dotsweep/SKILL.mdto anything reading it without a clone — the GitHub UI, a raw URL, a copy-paste install. The invariant was right and the mechanism was wrong.sync-skill.shwrites both copies and the validator compares them, which asserts the thing that actually matters. -
The version lives in two files —
metadata.versionin the skill frontmatter andversionin.claude-plugin/plugin.json— and they must move together.marketplace.jsondeliberately carries none, soplugin.jsonstays the single source of truth.scripts/bump-version.pycarries it across on sync so it is not a step someone has to remember. -
No machine-specific paths and no API host but
dotsweep.comin the prompt. Both were violated once in the commit that added the check for them.
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.
- 4d ago First seen · 52 lines · 640 tokens per session scan A 0583501c77f1
dotsweep CLAUDE.md is an instructions file published in the GitHub repository stanmaygo/dotsweep (3 stars, last pushed 1mo ago), licensed MIT. It adds 640 tokens to every session, about $0.0032 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to dotsweep AGENTS.md, differing in 0 lines, and is treated as a copy.
Other instructions, from other repositories
gortex CLAUDE.md
Claude Code instructions for zzet/gortex, covering gortex, build & test, codebase overview, discovery (read once, then keep using) and llm provider (powers ask and searchsymbols assist: modes).
agent-toolkit AGENTS.md
AGENTS.md instructions for ulises-jeremias/agent-toolkit, covering agents.md — ai agent contract, what this toolkit does, repository structure, operating rules and how to add a skill.
core-ai CLAUDE.md
Claude Code instructions for helius-labs/core-ai, covering core ai, compiler-managed sync, mcp server pin, skill versioning and router surface maintenance.
core-ai AGENTS.md
AGENTS.md instructions for helius-labs/core-ai, covering core ai — agent instructions, repository overview, mcp server setup, api key setup and public tool surface.
ChatCrystal AGENTS.md
AGENTS.md instructions for ZengLiangYi/ChatCrystal, covering agents.md, project overview, commands, development (server port 3721 + client port 13721) and build for production.
cursor-studio copilot-instructions.md
Copilot instructions for DearLicy/cursor-studio, a project described as: Local-first Windows control center for Cursor: manage AI providers, balances, usage and costs, sessions, prompts, MCP, Skills, appearance, and integrations. | Cursor 本地桌面管理工具.