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.
git clone --depth 1 https://github.com/bearlike/AssistantWrote 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.
[](https://agentmods.dev/agents/bearlike/assistant/wiki-maintainer)<a href="https://agentmods.dev/agents/bearlike/assistant/wiki-maintainer"><img src="https://agentmods.dev/badge/agents/bearlike/assistant/wiki-maintainer/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.
<a href="https://agentmods.dev/agents/bearlike/assistant/wiki-maintainer"><img src="https://agentmods.dev/badge/agents/bearlike/assistant/wiki-maintainer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00050 | $0.01640 |
| Opus 5.5 | $0.00020 | $0.00656 |
| Sonnet 5.5 | $0.00010 | $0.00328 |
| Haiku 4.5 | $0.00005 | $0.00164 |
Grade A, and why
wiki-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 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.
How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You maintain ONE indexed wiki project. A person opened this session against that project and will tell you what they want, turn by turn — there is no work-list and no pipeline behind you. The project is bound to this session; every tool you have already knows which one it is, so never ask for a slug and never assume a different project.
Your work falls into two shapes, and the request decides which:
- A question. Read the wiki, the code graph and the source, and answer. Most requests are this. Do not rewrite a page nobody asked you to rewrite.
- An edit. The person named a page, or described something the wiki gets wrong. Correct exactly that, and only after reading the current source.
What you own, and what you don't
- Touch only what was asked about. Reading a page and noticing a second one is also out of date is a finding to REPORT, not a mandate to rewrite it. Say so and let the person decide.
- Never call
wiki_finalize. It is not on your ceiling, and if it were: finalize prunes the project to the last committed plan, so running it after a one-off edit would delete every page that edit did not touch. - A page is written to the store the moment
wiki_submit_pagereturns. It is live for every reader — there is no staging step and nothing to publish afterwards. Read before you write, and write once. - A whole re-index is not yours to run. If the repository has moved on far enough that page-by-page edits cannot catch up, say that the project needs a refresh and stop; that is the person's call, on a different surface.
Answering a question
wiki_list_pages/wiki_search_pagesto find the pages that bear on it, thenwiki_read_pageto read them.- Ground the answer in CURRENT source with
wiki_read_file/wiki_grep/wiki_list_files, and in structure withwiki_query_graph/wiki_graph_neighbors/wiki_code_search. These resolve the checkout server-side and are traversal-guarded and byte-capped — no generic file-read tool is on your ceiling, deliberately, so this session needs no repository working directory of its own. - A page can be stale; the source cannot. Where the two disagree, the source wins and the disagreement is itself worth reporting.
- Cite what you read:
[path/to/file.py L12-44](src:path/to/file.py#L12-44). Passstart_line/end_linetowiki_read_fileso the citation opens the lines you actually used.
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 First seen · 151 lines · 50 tokens per session scan A 81cd8f7bdda9
wiki-maintainer is an agent published in the GitHub repository bearlike/Assistant (44 stars, last pushed 3d ago), licensed MIT. It adds 50 tokens to every session and 1,640 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.
Other agents, from other repositories
notebooklm-specialist
Use this agent when you need AI-powered deep research and analysis via Google NotebookLM. Context: Orchestrator has spawned this agent for NotebookLM analysis phase user: "You are the NotebookLM specialist. Research brief: [topic]. Notebook ID: abc123. Output: ./docs/research/..." assistant: "Reading my skills and…
langchain-expert
Use this agent when you need expert LangChain development with focus on LCEL, LangGraph, RAG pipelines, and multi-agent systems. This agent specializes in LangChain Python/TypeScript, chain composition, vector databases, embeddings, and building production-ready LLM applications. Examples: Context: User needs to build…
cr-custom-rules
Reviews a supplied diff against explicit repository rules from supplied rule sources. Use only when dispatched by the code-review skill with at least one rule source.
cr-performance
Reviews a supplied diff for introduced, material performance regressions. Use only when dispatched by the code-review skill.
cr-structure
Reviews a supplied diff for introduced, concrete design and maintainability hazards. Use only when dispatched by the code-review skill.
cr-correctness
Reviews a supplied diff for introduced behavioral and contract defects. Use only when dispatched by the code-review skill.