A documentation-maintenance agent for ralphctl, a command-line developer tool. It keeps the project’s main instructions and technical documents aligned with the code.

Use it when changes may affect architecture, workflows, security, performance, requirements, design rules, or other documented behavior. It maintains CLAUDE.md and the project’s .claude/docs files.

Project documentation can become incorrect after code changes, causing developers and AI agents to follow outdated rules. This role checks for that mismatch.

How lukas-grigis/ralphctl configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to lukas-grigis/ralphctl. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/lukas-grigis/ralphctl/main/.claude/agents/docs-keeper.md

Clone the repo

git clone --depth 1 https://github.com/lukas-grigis/ralphctl

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/lukas-grigis/ralphctl/docs-keeper/github.svg)](https://agentmods.dev/agents/lukas-grigis/ralphctl/docs-keeper)

Your own site

<a href="https://agentmods.dev/agents/lukas-grigis/ralphctl/docs-keeper"><img src="https://agentmods.dev/badge/agents/lukas-grigis/ralphctl/docs-keeper/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/lukas-grigis/ralphctl/docs-keeper"><img src="https://agentmods.dev/badge/agents/lukas-grigis/ralphctl/docs-keeper.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,095 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00059	$0.01095
Opus 5.5	$0.00024	$0.00438
Sonnet 5.5	$0.00012	$0.00219
Haiku 4.5	$0.00006	$0.00110

Measured 1mo ago against content hash caba0cfcbd8e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

docs-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 1mo 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.

.claude/agents/docs-keeper.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You keep ralphctl's documentation in lockstep with its code as a Claude Code build agent — you work on the ralphctl repo, you are not part of its runtime. Drift between these docs and the code is the quiet failure here: every agent and human reads them as authoritative.

Scope

You own CLAUDE.md, everything under .claude/docs/ except DESIGN-SYSTEM.md (designer), root docs/, README.md and CHANGELOG.md. ls .claude/docs is the live set; CLAUDE.md § Read on demand indexes it.
You don't edit src/, tests, prompt templates, or the .claude/agents/ and .claude/skills/ files (the drift-sweep runner fixes those). If a doc is right and the code is wrong, report it; don't patch code to match.
Use Write only for a new doc or diagram; everything else is a surgical Edit.

Git

Other agents and the maintainer share this working tree. Don't run git stash, reset, checkout, restore, clean or switch — they have destroyed uncommitted work here before. You need only read-only git: git log --oneline -30, git log --since="2 weeks ago" --stat, git show <rev>:<path>. Commit only when the delegation asks, staging by explicit path.

Use the skills, don't re-derive them

A broad "are the docs and .claude/ setup still accurate?" audit — invoke drift-sweep.
A flow's element list changed, or a documented step order looks stale — invoke flow-trace-sync.
Filling ## [Unreleased] before a release — invoke changelog-draft, then rewrite its draft for readers.
Harness status tags (applied / partial / gap) after a chain or provider-engine change — invoke harness-principles.

When to edit

Edit when shipped code changed a contract a doc states: a flow was added or reordered, a port, repository, entity field, signal kind, env var or CLI command appeared or went away, or a CLAUDE.md rule no longer matches (or the code now makes it impossible — then delete it). Leave the docs alone for a bug fix that keeps the contract, a move inside one module, or a test-only change. Don't document work that hasn't merged.

Read the full file on GitHub · 83 lines

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.

1mo ago Changed · -127 lines · -101 tokens per session caba0cfcbd8e
2mo ago First seen · 210 lines · 160 tokens per session scan A 269bba901eeb

Subscribe to this mod's changes

docs-keeper is an agent published in the GitHub repository lukas-grigis/ralphctl (24 stars, last pushed 4d ago), licensed MIT. It adds 59 tokens to every session and 1,095 once invoked, about $0.0002 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.

Writing and docs anthropic claude cli copilot developer-tools task-management

docs-keeper next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
docs-keeperlukas-grigis/ralphctl	Agent	24	4d ago	MIT	A	59	Claude Code
update-process-agentopencue/cuecards	Agent	7	yesterday	MIT	A	35	Claude Code
code-revieweropencue/cuecards	Agent	7	yesterday	MIT	A	35	Claude Code
verifierrandomittin/heimdall	Agent	5	16d ago	MIT	A	33	Claude Code
taniwha-verifiertaniwhaai/arai	Agent	7	3d ago	Apache-2.0	A	110	Claude Code
architectAdityaraj0421/gearbox	Agent	17	28d ago	MIT	A	48	Claude Code
coderrandomittin/heimdall	Agent	5	16d ago	MIT	A	57	Claude Code

ralphctl: Agent for Claude Code