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/romerma/mstack/claude-mdgit clone --depth 1 https://github.com/romerma/mstackWhat 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 | $0.00284 | $0.00284 |
| Opus 5 | $0.00142 | $0.00142 |
| Sonnet 5 | $0.00057 | $0.00057 |
| Haiku 4.5 | $0.00028 | $0.00028 |
Grade A, and why
mstack 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
mstack
This repository is the mstack plugin, and it uses mstack on itself: .mstack/ is the real
work queue for real work on the plugin, which is the only way an unused workflow gets found
out.
Workflow
Start work with /mstack. Durable state is in .mstack/; mstack gate must be green
before a session closes. One item active at a time, a verdict from a pass that did not
write the code to close it, and product forks answered through mstack decide --resolves.
Development
The dev loop, the runtime rules (no build step, no runtime dependencies, both runtimes
green), and the docs discipline (pasted output from real runs) are in
CONTRIBUTING.md. Run the plugin from this checkout with
claude --plugin-dir ., with the marketplace-installed copy disabled — otherwise the hooks
load twice and, worse, subagents run the cached agent and skill definitions rather than
this checkout's.
Every CLI call in this repository is ./bin/mstack, never bare mstack: the copy on PATH
is the installed plugin, which can be older than this checkout and reports green on checks
it does not contain. ./bin/mstack version prints which copy is running.
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 · 25 lines · 284 tokens per session scan A 0917d9cb2a74
mstack CLAUDE.md is an instructions file published in the GitHub repository romerma/mstack (1 stars, last pushed 8d ago), licensed MIT. It adds 284 tokens to every session, about $0.0014 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
spec-kitty AGENTS.md
Instructions for Priivacy-ai/spec-kitty, covering spec kitty development guidelines, ⚠️ critical: load the project charter first, ⚠️ critical: template source location, ⚠️ critical: use canonical sources, never improvise and ⚠️ critical: git workflow — no direct pushes to origin/main.
autospec AGENTS.md
Instructions for ariel-frischer/autospec, covering agents.md, prerequisites, commands, build & dev and single test.
gangsta AGENTS.md
Instructions for kucherenko/gangsta, covering contributing to gangsta, what this is, what will not be accepted, skill changes and the laws apply to contributors too.
featherspec memory-bank.instructions.md
FeatherSpec Memory Bank writing rules (thin loader).
gangsta GEMINI.md
Instructions for kucherenko/gangsta, a project described as: AI agentic skills framework for spec-driven development, built on the organizational model of mafia.
hashloom CLAUDE.md
Claude Code instructions for davet47/hashloom, covering working rules for hashloom, git is the user's, scope discipline, hashloom develops on hashloom and definition of done: >5x token reduction.