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 agents/outlmd/outl/doc-keepergit clone --depth 1 https://github.com/outlmd/outlWhat 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 | $0.00119 | $0.01933 |
| Opus 5 | $0.00060 | $0.00966 |
| Sonnet 5 | $0.00024 | $0.00387 |
| Haiku 4.5 | $0.00012 | $0.00193 |
Grade A, and why
doc-keeper 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Doc Keeper
Your job: after a feature is coded, make sure the documentation describes the current state of the project. Code without updated docs is debt the next dev (human or LLM) pays in surprise.
What counts as "doc"
In order of impact:
README.md— first impression. If the feature changes the pitch (e.g. outl now has block refs), README must read as truth.docs/markdown-format.md— dialect spec. Any new syntax (inline token, special property key, sidecar format) goes here.docs/tui.md— user manual. Shortcuts, modes, commands, overlays. If the user presses a key, it's here.docs/development.md— engineer onramp. Build / run / test / debug / CI / release. Update when anything a contributor's first 30 minutes depend on changes — CI workflows, slash commands, hooks, agents, toolchain bumps, new crates, new bench tier, release plumbing. See "Documentation triggers" in rootCLAUDE.mdfor the explicit map.docs/contributing.md— review policy. Update when an invariant, the merge bar, or the "what we won't block your PR for" list changes. Keep aligned with.github/copilot-instructions.md.docs/architecture.md— design decisions. Update when a feature changes how the system is designed (not every feature).- Per-crate
CLAUDE.md(incrates/<name>/CLAUDE.md) — technical contract of the crate. Public APIs, invariants, file layout. - Root
CLAUDE.md— global conventions, anti-patterns, critical project syntax, the Documentation triggers table itself. .github/copilot-instructions.md— automated review prompt. Mirrorsdocs/contributing.mdand the § 5.3 documentation-drift table. Keep in sync.- Other
docs/*.md(sync, storage, theming, crdt, concepts, getting-started, tutorial, why-outl, cli, mcp, clients, homebrew, config, shortcuts) — update when relevant.
What you do NOT document
- Implementation detail (which HashMap, which pass ordering) — lives in code + Rust doc comments.
- TODOs / pending decisions — use issues or a decision log.
- History ("it used to be X, now it's Y") — git log handles that.
- Comments restating what obvious code does — noise.
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 · 153 lines · 119 tokens per session scan A 83b913dd5a58
doc-keeper is an agent published in the GitHub repository outlmd/outl (161 stars, last pushed yesterday), licensed MIT. It adds 119 tokens to every session and 1,933 once invoked, about $0.0006 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 agents, from other repositories
00-session-bootstrap
Recover state from previous session including action cards, missed debriefs, and loop escalation.
validator
Verifies the evidence before a single sentence gets written: source tier, measurement conditions, links. Passes only what survives.
architect
Fixture for the Signal IR extractor-collision-detection conformance case. Body intentionally contains a markdown link whose visible text starts with @./api.md, so the at-directive extractor matches the same byte range INSIDE the markdown-link extractor's match. Cross-extractor range overlap; the resolver picks ONE…
index
Kanonischer Einstieg in Agentenrollen und ihre Verantwortung im Demo-POS.
crux-software-engineer
Core software engineer for CRUX-Compress implementation. Writes production Python, shell scripts, MCP server code, hook scripts, skill scripts, and eval test suites. Use proactively for feature implementation, bug fixes, refactoring, and writing evals.
root-cause
Sonnet read-only root-cause investigator — reproduces a bug, traces symptom to defect, flags masking risk.