Borrowing it
Nothing to install: this file belongs to velhud/PatchBay. 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/velhud/PatchBay/main/AGENTS.mdgit clone --depth 1 https://github.com/velhud/PatchBayWrote 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/velhud/patchbay/agents-md)<a href="https://agentmods.dev/instructions/velhud/patchbay/agents-md"><img src="https://agentmods.dev/badge/instructions/velhud/patchbay/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/velhud/patchbay/agents-md"><img src="https://agentmods.dev/badge/instructions/velhud/patchbay/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 | $0.01543 | $0.01543 |
| Opus 5.5 | $0.00617 | $0.00617 |
| Sonnet 5.5 | $0.00309 | $0.00309 |
| Haiku 4.5 | $0.00154 | $0.00154 |
Grade A, and why
PatchBay 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 16d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Repository purpose and scope
PatchBay supports two complementary workflows:
- Manager/worker mode — ChatGPT Pro is the manager; local Codex workers investigate, implement, execute, review, and verify.
- Author Mode — ChatGPT Pro is the source author; PatchBay deterministically transports packages, validates paths and manifests, materializes isolated candidates, runs authorized checks, and returns evidence. Author Mode must not initialize or delegate to Codex workers.
This root file is manager/worker-first because most of the repository implements the MCP, worker, Hub, and Edge control planes. For all files under src/patchbay/authoring/ and tests/authoring/, the nested src/patchbay/authoring/AGENTS.md contract is authoritative where it is more specific.
Do not collapse the two workflows into one hidden routing layer. Shared Git/process/storage primitives may be reused, but implementation authority, startup, state, tools, and recovery boundaries must remain explicit.
Current model policy
Every delegated worker MUST explicitly use model='gpt-6-luna' with reasoning_effort='high' by default; xhigh or max is allowed. No other model family, older Luna, ambient default, fallback, or escalation is permitted. If Luna is unavailable, report the blocker. Apply this same rule to nested helpers and every handoff. The primary manager is separate from delegated workers; Author Mode has no delegated author or recovery worker.
Core engineering rules
- Do not add secrets, tokens, private paths, or machine identifiers.
- Preserve localhost-first and authenticated-public-access defaults.
- Keep read-only orientation available. Mark every mutating/destructive tool accurately.
- Validate repository paths, package paths, export destinations, object identities, and manifest inputs before side effects.
- Do not weaken package-path validation, maintenance locking, lease checks, unknown-execution fencing, supervisor-launch handling, JSON-RPC validation, safe exports, or honest CLI exit codes.
- Public evidence must stay bounded and sanitized. Private instance identity must not leak through qualification responses.
- Update code, schemas, tests, prompt surfaces, README/quickstart, and testing instructions together when public behavior 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.
- 16d ago Changed · -67 lines · -2,330 tokens per session 58bf24247ca3
- 2mo ago First seen · 165 lines · 3,873 tokens per session scan A d36478ba052e
PatchBay AGENTS.md is an instructions file published in the GitHub repository velhud/PatchBay (5 stars, last pushed 12d ago), licensed MIT. It adds 1,543 tokens to every session, about $0.0062 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 2026-08-31.
PatchBay AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| PatchBay AGENTS.mdvelhud/PatchBay | Instructions file | 5 | 12d ago | MIT | A | 1,543 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |