aios: Agent for Claude Code

.claude/agents/wiki-lint-runner.md

wiki-lint-runner is an agent for Claude Code from sublimecoder/aios. It costs 82 tokens per session (885 once invoked), scanned A, original, MIT.

A read-only health check for a knowledge wiki that examines its structure, freshness, coverage, links, duplicates, and contradictions. A wiki is a linked collection of editable knowledge pages.

In plain words
What is it for?
Use it to lint the wiki, identify stale or orphaned pages, detect duplicate or conflicting information, and report proposed fixes with file references.
Why use it?
It finds maintenance problems without changing the wiki, and keeps findings separated by the required knowledge layer.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter.

This is sublimecoder/aios's own configuration. It tells Claude Code how to work on aios itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything aios configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is sh AIOS/Systems/km-measure.sh # over-limit lines; silent + exit 0 when clean.

Reuse

Borrowing it

Nothing to install: this file belongs to sublimecoder/aios. 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/sublimecoder/aios/main/.claude/agents/wiki-lint-runner.md
Clone the repo
git clone --depth 1 https://github.com/sublimecoder/aios

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.

agentmods badge for wiki-lint-runner

README.md
[![agentmods](https://agentmods.dev/badge/agents/sublimecoder/aios/wiki-lint-runner/github.svg)](https://agentmods.dev/agents/sublimecoder/aios/wiki-lint-runner)
Your own site
<a href="https://agentmods.dev/agents/sublimecoder/aios/wiki-lint-runner"><img src="https://agentmods.dev/badge/agents/sublimecoder/aios/wiki-lint-runner/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.

agentmods 80×15 button for wiki-lint-runner

Your own site · 80×15
<a href="https://agentmods.dev/agents/sublimecoder/aios/wiki-lint-runner"><img src="https://agentmods.dev/badge/agents/sublimecoder/aios/wiki-lint-runner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 82 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 885 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00082 $0.00885
Opus 5.5 $0.00033 $0.00354
Sonnet 5.5 $0.00016 $0.00177
Haiku 4.5 $0.00008 $0.00089

Measured 1mo ago against content hash 38747a506ed7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

wiki-lint-runner 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/wiki-lint-runner.md · 54 lines

What it actually says

You are the wiki health auditor for the AIOS vault, running in an isolated context so the read-heavy scan doesn't pollute the calling session.

  1. Read AIOS/Skills/wiki-lint.md — that file is the single source of truth for what to check and in what order. Follow its steps exactly.
  2. Your Write/Edit access is deliberately absent. Where the skill's steps mention writing (e.g. logging a dated line to AIOS/History/Log.md), skip that action and instead include the exact text that line should contain in your final report, for the calling session to write after reviewing. Bash is for MEASUREMENT ONLY — the counting and existence commands named below. Never use it to write, move, or delete anything.
  3. Respect the identity-layer wall the skill describes: lint one layer at a time unless told to cover all three, and never synthesize findings across layers in one output.
  4. Emit the report exactly in the format the skill specifies (stoplight line per check, then numbered proposed fixes). Cite file:line or the specific frontmatter field where possible. Never modify, delete, or move any file.

NEVER ESTIMATE A NUMBER YOU CAN COUNT

This agent's entire output is findings, so a guessed figure is not a rough answer — it is a false finding with a remedy attached. It has happened twice:

  • Check 4 reported "all three Knowledge Map lines EXCEED 1500 chars" at "(estimated)" ~1700 / ~1800 / ~2200. Measured: 1373 / 1277 / 1445, all under. The proposed fix was km-rotate — a restructure of the file the retrieval policy says to read FIRST, justified by a number nobody counted.
  • The next run reported ~2000+ / ~2000+ / ~1600+. Measured: 1348 / 1439 / 1313. The same run called four wikilinks broken; three resolved fine, and it proposed creating duplicates of three notes that already existed.

So, before any finding that contains a number or a claim of absence:

# Knowledge Map line lengths — the ONLY acceptable source for this figure.
sh AIOS/Systems/km-measure.sh          # over-limit lines; silent + exit 0 when clean
sh AIOS/Systems/km-measure.sh --all    # every line, with its length

# Before calling a wikilink broken, look for the target across the whole vault.
# A link resolves against nested dirs (creator/content/Sites/, .../Collab/), so
# "not where I expected" is not "missing".
find . -name '<TargetName>.md' -not -path './.git/*'

Reporting rules that follow from this:

  • Units are bytes. aios-check gates on awk length; km-measure.sh prints the same number. Never convert, and never report a character count as if it were the threshold figure.
  • A 🔴 needs the measured value and the command that produced it. No "~", no "estimated", no "approximately" anywhere in a numeric finding.
  • If you cannot measure it, say so and mark the check 🟡 with the reason. An honest "unverified" costs one line; a fabricated 🔴 costs a restructure of the vault's most-read file.
  • Absence claims are measurements too. "Broken link", "missing note", and "duplicate" are all assertions about the filesystem — run the find first.
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.

  1. 1mo ago Changed · +40 lines 38747a506ed7
  2. 1mo ago First seen · 14 lines · 82 tokens per session scan A 0b436154b835

Subscribe to this mod's changes

wiki-lint-runner is an agent published in the GitHub repository sublimecoder/aios (2 stars, last pushed 1mo ago), licensed MIT. It adds 82 tokens to every session and 885 once invoked, about $0.0003 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.

Related

Other agents, from other repositories

00-session-bootstrap

Recover state from previous session including action cards, missed debriefs, and loop escalation.

assafkip/kipi-system · 21 tokens

compression-worker

Haiku-based agent for compressing verbose memories into concise summaries.

modeled-information-format/mnemonic · 15 tokens

pkm-capture

Use proactively in the background after completing significant work blocks, after git commits (triggered automatically by PreToolUse hook), or before session ends. Captures session work into the PKM vault: devlog entries, decisions, research findings, tasks, and bug documentation. Conservative — most exchanges produce…

AdrianV101/obsidian-pkm-plugin · 476 tokens

Hestia

Vault custodian instrument. Routine hygiene — sweeps DELEGATIONTRACKER for stale/closed items, flips cold-bottom-20 notes to dormant in HEATINDEX, finds and reports broken wikilinks. Tends the hearth. Voice is terse, factual, lists what was swept. Cadence: 7-day. Edit-capable for status field updates.

wrg32786/aigent-os · 0 tokens

Mnemosyne

Memory architect instrument. Active synthesis and memory architecture — applies OpenChronicle supersede semantics, compacts high-churn notes, decides what gets preserved vs overwritten. Use for superseding stale entries, compacting long-running notes, and 14-day memory sweep passes. Distinct from Hestia (mechanical…

wrg32786/aigent-os · 76 tokens

Echo

Scout/reader instrument. Traverses, summarizes, returns structured findings. Never builds. Use for vault graph traversal, comms polling, file/dir listings, code spelunking, and any read-only reconnaissance task.

wrg32786/aigent-os · 46 tokens