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/william-london/ownframework-loop/agents-mdgit clone --depth 1 https://github.com/william-london/ownframework-loopWhat 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 | $0.01569 | $0.01569 |
| Opus 5 | $0.00785 | $0.00785 |
| Sonnet 5 | $0.00314 | $0.00314 |
| Haiku 4.5 | $0.00157 | $0.00157 |
Grade A, and why
ownframework-loop 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 2d 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 — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This is the repository-local operating contract for coding agents working on
ownframework-loop. It is public, self-contained doctrine for this repository;
it does not depend on private infrastructure or private policy files.
Product intent
OwnFramework Loop is a deterministic engineering protocol for AI coding agents. The normal operator experience is deliberately low-friction:
spec → supervisor enqueue → autonomous build/review/repair → terminal result → operator promotion
There is no mandatory approval ceremony, confirmation token, manual PROGRAM initialization, manual claim/finalize sequence, or checkpoint babysitting.
The first legitimate execution start creates an immutable execution seal that binds the exact packet and spec-time source baseline. Promotion and external effects remain outside Loop authority.
Scope
This contract covers this repository only, including:
- deterministic source under
lib/,bin/,schemas/,templates/; - tests and validation under
tests/,validate.sh,release_gate.sh; - Claude integration under
skills/,agents/,hooks/,.claude-plugin/; - portable host-neutral skills under
.agents/skills/; - adapter metadata/docs under
adapters/anddocs/architecture/; - public project surfaces such as
README.md,SECURITY.md,CHANGELOG.md; - CI under
.github/workflows/.
Core versus adapter authority
The deterministic core owns:
- work-packet parsing and validation;
- spec-time baseline capture;
- first-start execution sealing;
- lifecycle transitions and locking;
- scope/runtime/repair budgets;
- candidate branch/worktree identity;
- exact candidate Git SHA;
- build receipts and exact-SHA verdict binding;
- crash reconciliation and terminal semantics;
- the boundary before operator promotion.
Adapters may provide host-specific skills, agents, hooks, discovery, and foreground/debug UX. Platform service installers and the core runtime remain OwnFramework Loop-owned. Adapters must not create a parallel execution-seal, state, repair, candidate, verdict, runtime-generation, scheduler, or promotion truth.
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.
- 2d ago First seen · 203 lines · 1,569 tokens per session scan A 0d9f1cafc799
ownframework-loop AGENTS.md is an instructions file published in the GitHub repository william-london/ownframework-loop (0 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,569 tokens to every session, about $0.0078 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
neat CLAUDE.md
Instructions for neat-technologies/neat, covering claude.md, binding rules, what neat is, what success looks like and conventions.
legal-skills AGENTS.md
Instructions for pa1nrui1/legal-skills, covering repository instructions and legal formal deliverable hard gate.
guide CLAUDE.md
Instructions for rios0rios0/guide, covering claude.md, project overview, build, test, and validate, update-wiki tool (go 1.26.2) — syncs docs to github wiki and toc sync validation (run from repo root).
mdxg-redline AGENTS.md
Instructions for oubakiou/mdxg-redline, a project described as: Inline review tool for MDXG markdown — capture highlighted comments and export them as structured JSON for LLM agents.
security-knowledge-base CLAUDE.md
Instructions for dinosn/security-knowledge-base, a project described as: Local-first, evidence-led security knowledge base with a model-neutral JSON CLI, immutable sources, and human-reviewed proposals.
lantern AGENTS.md
Instructions for lanseqingling/lantern, covering agents.md, project, commands, repository guide and architecture boundaries.