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/sno-ai/mda/claude-mdgit clone --depth 1 https://github.com/sno-ai/mdaWrote 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/sno-ai/mda/claude-md)<a href="https://agentmods.dev/instructions/sno-ai/mda/claude-md"><img src="https://agentmods.dev/badge/instructions/sno-ai/mda/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 5.1 | $0.01248 | $0.01248 |
| Opus 5 | $0.00624 | $0.00624 |
| Sonnet 5 | $0.00250 | $0.00250 |
| Haiku 4.5 | $0.00125 | $0.00125 |
Grade A, and why
mda 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 6d 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — sno-ai/mda
Project memory for Claude Code. Read on every session.
Release process for @markdown-ai/cli
No more asking the user about version numbers, npm tokens, trusted publishing, or auth. Everything below is settled. Just execute.
To cut a release (any 1.1.x, 1.2.0, 2.0.0 — same flow)
- Edit
apps/cli/package.json→ bumpversion pnpm cli:test(must pass; tests dynamically read the version)git add apps/cli/package.json && git commit -m "chore(cli): bump to <version>"git push origin maingit tag v<version> HEADgit push origin v<version>gh run watch $(gh run list --workflow=publish-cli.yml --limit 1 --json databaseId --jq '.[0].databaseId')npm view @markdown-ai/cli@<version> versionto confirm registry- Write real GitHub Release notes by hand and publish them. See "Release notes — non-negotiable" below. This step is part of the release; the release is not done until the GitHub Release exists with a written body.
The user does not need to do anything for any of these steps. Do not ask which version number to use — for a publish-only request, increment the patch. Do not ask whether to write release notes — always write them.
Release notes — non-negotiable
The GitHub Release body must be hand-written for every version.
--generate-notes produces a useless "Full Changelog: vA...vB" one-liner
when there are no merged PRs between tags (which is most patch releases
in this repo), so the workflow no longer creates the Release
automatically. Writing the body is the agent's responsibility.
Minimum required structure:
## What's new
<one sentence per user-facing change, lead with the change name in bold.
example: **`mda demo`** — one command to see the four-artifact flow ...>
## Other changes
<bullet list of smaller things: docs, CI, internal>
## Why the version jump (only if patches were skipped)
<one paragraph explaining which version numbers were skipped and why>
## Install / upgrade
<copy-pasteable npm install + npx invocation>
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.
- 6d ago First seen · 120 lines · 1,248 tokens per session scan A f4101495430b
mda CLAUDE.md is an instructions file published in the GitHub repository sno-ai/mda (616 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 1,248 tokens to every session, about $0.0062 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
EverOS CLAUDE.md
Claude Code instructions for EverMind-AI/EverOS, covering everos — md-first memory extraction framework, quick commands, architecture, engineering practices and branch strategy.
llm-wiki CLAUDE.md
Instructions for nvk/llm-wiki, covering llm-wiki development guide, testing, github auth and transport, structural tests (always run — no llm, instant) and behavioral evals (run when changing command logic).
relay AGENTS.md
AGENTS.md instructions for AgentWorkforce/relay, covering git workflow rules, never push directly to main, correct workflow, ... do work .. and stop here - let user merge.
disp8ch AGENTS.md
AGENTS.md instructions for aaronnat23/disp8ch, covering agents, repo instructions and session startup.
huf CLAUDE.md
Claude Code instructions for tridz-dev/huf, covering claude.md, project overview, repository structure, tech stack and backend.
ai-job-finder CLAUDE.md
Instructions for chrisdothtml/ai-job-finder: This repo uses Typescript with node.js, with Yarn as the dependency manager. You can use ./bin/tsn as a drop-in replacement for the node binary. I highly prefer to avoid adding additional third-party dependencies to my project, unless it abstracts very complex…