Borrowing it
Nothing to install: this file belongs to btspoony/mstar-harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/btspoony/mstar-harness/main/AGENTS.mdgit clone --depth 1 https://github.com/btspoony/mstar-harnessWrote 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/btspoony/mstar-harness/agents-md)<a href="https://agentmods.dev/instructions/btspoony/mstar-harness/agents-md"><img src="https://agentmods.dev/badge/instructions/btspoony/mstar-harness/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/btspoony/mstar-harness/agents-md"><img src="https://agentmods.dev/badge/instructions/btspoony/mstar-harness/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.02245 | $0.02245 |
| Opus 5.5 | $0.00898 | $0.00898 |
| Sonnet 5.5 | $0.00449 | $0.00449 |
| Haiku 4.5 | $0.00225 | $0.00225 |
Grade A, and why
mstar-harness 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 7d 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Morning Star Harness Maintenance
This repository contains the Morning Star runtime skills, workflow engine, CLI, and host plugins. This file defines repository maintenance policy; runtime behavior belongs in skills/mstar-*.
Working rules
- Read the affected source and direct contracts before editing. Fix the demonstrated problem or requested outcome with the smallest complete change; avoid speculative features, compatibility layers, and unrelated refactors.
- Keep each rule in its authoritative topic. Follow harness core for runtime semantics and roles for skill selection; do not duplicate their load matrices, gates, or role procedures here.
- Use tools actually available in the session. Continue authorized work through verification; ask only when a material decision or required authorization remains unresolved. Do not add a separate human approval gate for every diff.
- Respect existing assignments and write ownership. Preserve other contributors' changes; do not reset or clean their work.
- Do not edit user secrets or credential files. Changes to global host configuration or security-sensitive defaults require explicit user authorization.
- No hard block without a way out. Every refusal a user-facing entry can emit must name the actual cause and state the exact recovery the operator can perform through a supported, documented interface — the same entry or another documented one, but never a hidden or manual escape (hand-editing state, undocumented flags, or steps the tooling does not expose). A refusal that leaves the operator with nothing to do, a gate with no paired recovery, and a generic "blocked" message that masks an identifiable cause are all defects. Never add a hard limit without shipping its recovery in the same change, and cover both with a test.
- Fault tolerance and discoverability first for agent-facing operations. All document and content mutations are agent-authored, and agent mistakes are expected, not exceptional. Before any restriction: mutations stay revisable through ordinary public operations (no one-way doors, no hidden manual repair), every parameter is optional except a minimal
requiredset with safe defaults, and commands, parameters, and recovery paths are discoverable from--helpand refusal text — an undiscoverable capability does not exist. Principle anchor and audit backlog: #365. - Test layout. Module-unit suites colocate as
src/<module>.test.ts(unit mirror of one module); cross-module behavior suites live intest/. Test-support helpers live with the package that owns them. Engine note: the engine build compilessrc/**, so colocated suites are typechecked and their declarations land indist/. - Tests derive volatile facts, never pin them. A test that asserts a repository fact — the release version, an inventory of packages/roles/skills or its count, a checkout path — MUST read it from the source that owns it (the manifest, a directory listing, the CLI). A literal is acceptable only as a synthetic fixture input that nothing compares against reality:
"4.0.0"as input is fine,version === "4.0.0"is not. This class has been cleaned up across the engine, dsh and CLI suites before and came back, because a one-off cleanup leaves no rule behind. A pinned release version reddens every release branch whilemainstays green, so ordinary PR CI cannot surface it — derive the value, or the failure lands on whoever prepares the next release.
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.
- 7d ago Changed · +1 lines · +170 tokens per session bec7cc7d5bf2
- 8d ago Changed · +67 tokens per session 8acf2564c519
- 17d ago Changed · +27 tokens per session 80b741a31369
- 21d ago Changed · +1 lines · +109 tokens per session eaa64d9ddfc0
- 22d ago Changed · +1 lines · +127 tokens per session bd751aa228cc
- 23d ago Changed · +1 lines · +76 tokens per session 8556fa9227db
- 1mo ago Changed · +1 lines · +109 tokens per session 084a813cd19f
- 1mo ago Changed · -193 lines · -2,877 tokens per session 2fa0fcd880f0
- 1mo ago Changed · +1 lines · +66 tokens per session 45a360a17209
- 1mo ago First seen · 257 lines · 4,371 tokens per session scan A a08af9e6495a
mstar-harness AGENTS.md is an instructions file published in the GitHub repository btspoony/mstar-harness (64 stars, last pushed yesterday), licensed MIT. It adds 2,245 tokens to every session, about $0.0090 per session on Opus 5.5. 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).