LLM Wiki is a system that uses AI agents to research sources, compile knowledge bases, organize project briefs, and answer questions from the resulting wiki. It is used to turn rough ideas and investigations into structured, reusable knowledge for agents, including Claude Code, Codex, and OpenCode. The catalogue entries are commands, hooks, skills, instructions, and a plugin that implement its workflow.
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 commands/nvk/llm-wiki/ingest-collectiongit clone --depth 1 https://github.com/nvk/llm-wikiWrote 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/commands/nvk/llm-wiki/ingest-collection)<a href="https://agentmods.dev/commands/nvk/llm-wiki/ingest-collection"><img src="https://agentmods.dev/badge/commands/nvk/llm-wiki/ingest-collection.svg" alt="Measured on agentmods" 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 | $0.00031 | $0.03796 |
| Opus 5 | $0.00015 | $0.01898 |
| Sonnet 5 | $0.00006 | $0.00759 |
| Haiku 4.5 | $0.00003 | $0.00380 |
Grade A, and why
ingest-collection scanned grade A with 1 finding 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 6d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
allowed-tools: Read, Write, Edit, Glob, Grep, Bash(ls:*), Bash(wc:*), Bash(date:*), Bash(mkdir:*), Bash(mv:*), Bash(cp:*), Bash(rm:*), Bash(basename:*), Bash(find:*), Bash(git:*), Bash(curl:*), Bash(python3:*), Bash(bunz How it starts
The opening of the file, as written. The whole thing — 260 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Your task
Bulk-ingest a collection into the wiki as immutable raw sources. A collection is a bounded upstream corpus, not a single article: a Git repository full of specs, a BIP repository, a MediaWiki XML dump/API site, a CSV/JSON message archive, or a Wayback CDX snapshot set. Do not compile one wiki article per upstream page by default. First preserve raw sources with provenance, then compile synthesized topic/concept/reference articles.
Resolve the wiki. Follow the same resolution flow as /wiki:ingest:
- Read
$HOME/.config/llm-wiki/config.json. If it hashub_path, expand leading~only and prefer that path; useresolved_pathonly as a fallback cache when the expandedhub_pathis unavailable andresolved_pathis initialized. If config has onlyresolved_path, use it. If the configured path can be statted but readingwikis.jsonor listingtopics/fails withOperation not permitted, stop and ask the user to grant Full Disk Access/iCloud Drive access to the launcher; do not fall back to~/wikiorresolved_path. Do not write machine-specificresolved_pathinto shared configs. - If no config → read
$HOME/wiki/_index.md. If it exists → HUB =$HOME/wiki. If nothing found, ask the user where to create the wiki. - Wiki location, first match:
--local→.wiki/;--wiki <name>→HUB/wikis.jsonlookup with portable path resolution (<HUB>,~, absolute, or HUB-relative); if the registry path is stale, fall back toHUB/topics/<name>; current directory has.wiki/→ use it; else → HUB. - If
<wiki>/_index.mdis missing and--new-topic <name>is set, create the topic wiki using the init protocol before ingesting. If no wiki exists and no--new-topic, stop and ask for a target wiki.
Archive rule: collection ingest skips archived topic wikis by default. If
--wiki <name> resolves to status: archived or a path under
topics/.archive/, stop and ask the user to restore it with
/wiki:archive restore <name> or rerun with --include-archived. When
explicitly included, write only inside that archived topic path and keep it
archived. Auto-classification and --new-topic collision checks should treat
archived topics as unavailable unless the user explicitly restores or includes
archived context.
Read skills/wiki-manager/references/ingestion.md and skills/wiki-manager/references/wiki-structure.md, then follow the collection ingestion protocol.
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.
- 6d ago First seen · 260 lines · 31 tokens per session scan A 24566d0080a4
ingest-collection is a command published in the GitHub repository nvk/llm-wiki (1,193 stars, last pushed 9d ago), licensed MIT. It adds 31 tokens to every session and 3,796 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
orchestrate
This tool coordinates Claude Code's actions. It does NOT write code or create content.
project
Generate project documentation (product.md, structure.md, tech.md, codemaps/).
spreadsheet-audit
Perform a 6-step audit of an Excel spreadsheet for formula correctness, formatting consistency, data integrity, boss-auditable simplicity, and spec compliance. Uses the excel-screenshot MCP server tools — no Excel installation required.
performance-bottlenecks
Identify and resolve performance bottlenecks in your development workflow.
test-feature
Test a React Native feature on the running simulator/emulator. Verifies UI, user flows, and internal state. Generates a persistent Maestro test file.
ingest-l1
L1 analysis loop for the abapwiki knowledge base: for each batch it launches the abap-analyzer sub-agent in parallel, then the adversarial judge abap-deepcheck (separate session), applies only the analyses that pass the fail-closed gate, and commits. Resumes exactly after an interruption. Use this skill to document…