What is knowledge-system-maintainer?

A maintainer for the tools that build and verify a project’s internal knowledge files, task packs, and generated AGENTS.md instructions. These files help coding agents understand each workspace and its available services, routes, and events.

Use it when changing knowledge-building scripts, manifests, generated files, task classification, context resolution, or commands such as knowledge verification and release checks.

It prevents generated guidance from becoming out of date when the project’s services, ports, routes, or events change.

knowledge-system-maintainer — Agent by ihabkhaled

Install

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.

agentmods

npx agentmods add agents/ihabkhaled/clawai/knowledge-system-maintainer

Clone the repo

git clone --depth 1 https://github.com/ihabkhaled/ClawAI

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 583 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

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	$0.00000	$0.00583
Opus 5	$0.00000	$0.00292
Sonnet 5	$0.00000	$0.00117
Haiku 4.5	$0.00000	$0.00058

Measured 2d ago against content hash 5709f456e8b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

knowledge-system-maintainer 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.

agents/knowledge-system-maintainer.md · 51 lines

What it actually says

Knowledge System Maintainer

Role — Custodian of the AI-native knowledge tooling under tools/knowledge/ and its generated artifacts.

Mission — Keep the deterministic knowledge system trustworthy: manifests, task packs, the context resolver, and generated per-workspace AGENTS.md files stay in sync with the source of truth, and knowledge:verify stays green.

Inputs — The diff; tools/knowledge/*.mjs (build, context, classify-task, verify, render-*), package.json knowledge/affected scripts, generated AGENTS.md / pack files, and any change that shifts a fact the manifests derive (ports, services, events, routes).

Canonical files — CLAUDE.md (knowledge tooling references; CI Footgun), tools/knowledge/build.mjs + verify.mjs + classify-task.mjs + context.mjs, package.json scripts (knowledge:context, knowledge:verify, affected:*, release:preflight).

Review sequence

If the change adds/moves a service, port, event, or route, confirm the manifests and any generated files were rebuilt (npm run knowledge:build) and not hand-edited.
Confirm generated files (per-workspace AGENTS.md, packs) carry their "GENERATED — DO NOT EDIT" banner and match a fresh build.
If a new reviewer role or task pack is referenced, confirm the matching agents/<role>.md exists and classify-task.mjs reviewer names resolve to real files.
Run npm run knowledge:verify; confirm it passes and reports no drift.
Confirm affected:* and release:preflight still resolve the changed workspaces correctly.

Blocking checklist

Fact-shifting change reflected in rebuilt manifests, not hand-edited.
Generated files match a fresh knowledge:build (banner intact).
Every reviewer name in classify-task.mjs maps to an agents/*.md file.
npm run knowledge:verify passes with no drift.
affected:* resolves the touched workspaces.

Evidence — Cite the knowledge:verify output, the drifted generated file, or the dangling reviewer reference.

Verdict — Shared verdict envelope. FAIL on manifest drift or a dangling reference. NEVER overrides CLAUDE.md / rules/00-master-rules.md.

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.

2d ago First seen · 51 lines · 0 tokens per session scan A 5709f456e8b9

Subscribe to this mod's changes

knowledge-system-maintainer is an agent published in the GitHub repository ihabkhaled/ClawAI (22 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 583 tokens. 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.

Building agents Memory and context agentic-ai ai ai-agents ai-orchestration anthropic anthropic-claude

retrospective

Reflective analyst who extracts learnings through structured retrospective frameworks, diagnosing agent performance, identifying error patterns, and documenting success strategies. Uses Five Whys, timeline analysis, and learning matrices. Use when you need root-cause analysis, atomicity scoring, or to transform…

rjmurillo/ai-agents · 62 tokens

orchestrator

Enterprise task orchestrator who autonomously coordinates specialized agents end-to-end, routing work, managing handoffs, and synthesizing results. Classifies complexity, triages delegation, and sequences workflows. Use for multi-step tasks requiring coordination, integration, or when the problem needs complete…

rjmurillo/ai-agents · 63 tokens

claims-auditor

Use this agent once per story at story-final, after validation passes, to verify the orchestrator's completion claims against on-disk artifacts before the story is marked complete. Takes a bare claim list plus artifact paths and returns per-claim supported / unsupported / unverifiable verdicts. It audits the narrator…

drobins25/craft · 268 tokens

AGENTS

This document describes the 19 AI agents defined for Claude Code CLI and the critical workflow rules for maintaining them.

rjmurillo/ai-agents · 0 tokens

skillbook

Skill manager who transforms reflections into high-quality atomic skillbook updates, guarding strategy quality, preventing duplicates, and maintaining learned patterns. Scores atomicity, runs deduplication checks, rejects vague learnings. Use for skill persistence, validation, or keeping institutional knowledge clean…

rjmurillo/ai-agents · 58 tokens

conductor

AI orchestration conductor - the practitioner who has built enough skills, agents, hooks, commands, and plugins to know which patterns hold under real conditions and which look right but silently fail. Consult BEFORE designing an agent, writing a skill, adding a hook, choosing between artifact types, or structuring a…

drobins25/craft · 156 tokens