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/bbc/bug/agents-mdgit clone --depth 1 https://github.com/bbc/bugWhat 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 | $0.01284 | $0.01284 |
| Opus 5 | $0.00642 | $0.00642 |
| Sonnet 5 | $0.00257 | $0.00257 |
| Haiku 4.5 | $0.00128 | $0.00128 |
Grade A, and why
bug 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 today.
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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Agent Guidelines
Module Testing Standard
- Use docker-based test execution for module containers.
- For module container packages, set
scripts.testto build and run tests in Docker using each module'sDockerfile.test. - Prefer this pattern for module test commands:
docker build -f ./Dockerfile.test -t bug-module-test ../../../.. && docker run --rm --name bug-module-test-run bug-module-test npx jest --runInBand --config ./jest.config.cjs
Module Container Quick Check Profile
- Purpose: run a quick agentic pre-commit check in chat before major module container changes are committed.
- Use trigger phrase in chat: Run the module container quick check before commit.
Implementation Defaults
- Rule: keep worker restart checks aligned to real config dependencies.
- Rule: default to simple, readable implementations over maximum efficiency; only optimize for performance when there is a clear, measured requirement.
- Rule: prefer the simplest safe fix first, with minimal code changes.
- Rule: prefer inlined one-off literal values in module runtime files instead of file-level constants when the value is used once and naming does not add clarity.
- Rule: prefer no backward compatibility in new code; avoid aliases/shims/migration layers unless explicitly requested.
- Rule: never add regression tests for changes unless a task explicitly asks for them.
API and Service Patterns
- Rule: in module containers, prefer service-layer throw-on-error and keep routes thin; let express-async-handler propagate thrown/rejected errors via middleware instead of route-level
{ error }handling. - Rule: in module containers, services should return raw payloads (object/array/value), not
{ data: ... }; route handlers should wrap service results in API response envelopes. - Rule: status-check services are the exception to the general throw-on-error pattern; they should catch operational errors, log them, and return
[]instead of throwing.
Worker Architecture
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.
- today First seen · 73 lines · 1,284 tokens per session scan A 75765fdf4555
bug AGENTS.md is an instructions file published in the GitHub repository bbc/bug (43 stars, last pushed 7d ago), licensed Apache-2.0. It adds 1,284 tokens to every session, about $0.0064 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 2026-09-01.
Other instructions, from other repositories
sockudo AGENTS.md
Instructions for sockudo/sockudo, covering agents.md, mission, autonomy, repository map and core principles.
winx-code-agent AGENTS.md
Instructions for gabrielmaialva33/winx-code-agent, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
moclojer AGENTS.md
Instructions for moclojer/moclojer, covering agents.md, project overview, development commands, run the server and run all tests.
sockudo CLAUDE.md
Instructions for sockudo/sockudo, covering claude.md, mission, autonomy, repository map and core principles.
API-ARENA CLAUDE.md
Claude Code instructions for Eduu115/API-ARENA, covering api arena — claude code memory and project rules (canonical, shared with cursor).
moclojer copilot-instructions.md
Instructions for moclojer/moclojer, a project described as: Simple and efficient HTTP and Webscoket mock server with specification in yaml, edn or OpenAPI.