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 skills add nemock/company-brain --skill querygit clone --depth 1 https://github.com/nemock/company-brainWrote 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/skills/nemock/company-brain/query)<a href="https://agentmods.dev/skills/nemock/company-brain/query"><img src="https://agentmods.dev/badge/skills/nemock/company-brain/query/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/skills/nemock/company-brain/query"><img src="https://agentmods.dev/badge/skills/nemock/company-brain/query.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 | $0.00051 | $0.02097 |
| Opus 5.5 | $0.00020 | $0.00839 |
| Sonnet 5.5 | $0.00010 | $0.00419 |
| Haiku 4.5 | $0.00005 | $0.00210 |
Grade A, and why
query 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
query
This skill answers questions by reading the typed graph in a company-brain vault. You are the retrieval analyst; the vault is the source of truth.
The pattern follows the Infinite Brain retrieval analyst convention, adapted for a multi-product, multi-stakeholder company graph:
- Auto-inject the relevant pillars before reasoning about the user's question.
- Select candidate nodes by summary relevance, type, confidence, recency.
- Expand the candidate set by walking typed edges.
- Answer with node-id citations. Every claim names the node it comes from.
- Flag staleness, contradictions, and low confidence explicitly.
You do not invent facts. If the vault does not contain the answer, say so — and offer to capture the gap with the intake skill rather than guessing.
Before any question
Do these first when the skill is invoked:
- Confirm the vault path. Default: current working directory. Resolve to absolute. Refuse if the path has no
_system/PROFILE.md. - Load the active schema. Run:
The returned JSON tells you the active profile, controlled-document-footer policy, and the full list ofcb describe-profile --path <vault>active_node_types. The profile decides which node folders even exist. - Load the pillar set. Run:
This returns every pillar withcb list-nodes --path <vault> --auto-inject-onlyauto_inject: trueplus itsapplicable_whenstring. These are the governing principles of the company. Skim them up front. They will shape your answer even when the user's question is narrow.
Staged retrieval
Stage A — auto-inject relevant pillars
Each pillar carries an applicable_when field listing the topics it governs (e.g. "pricing, business model, pad, disposable, recurring revenue"). Match the user's question against applicable_when strings; load the body of any pillar that matches. These pillars are facts about how this company thinks — they govern the answer even when not explicitly cited by the user.
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 · 140 lines · 51 tokens per session scan A 1038a24145db
query is a skill published in the GitHub repository nemock/company-brain (5 stars, last pushed 4mo ago), licensed MIT. It adds 51 tokens to every session and 2,097 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 2026-08-31.
Other skills, from other repositories
agent-recall
Persistent compounding memory for AI agents. 5 default MCP tools: sessionstart, sessionend, remember, recall, check. Full surface: 6 tools with --full (adds checkaction); 13 tools with AREXTRAS=1 --full. Two-verb model: inhale (sessionstart) and exhale (sessionend). Correction-first memory with decision trail…
swarmvault
Use SwarmVault when the user needs a local-first knowledge vault that writes durable markdown, graph, search, dashboard, review, chat-session, context-pack, task-ledger, static AI export, retrieval, and MCP artifacts to disk from books, notes, transcripts, exports, datasets, slide decks, files, URLs, code, and…
linker
Discovers missing cross-domain connections between knowledge documents and creates/updates connection wiki articles.
extract-memory
Extract writing style, personality traits, and key insights from a collection of documents (Markdown, PDF, TXT, DOCX) and save them as a Claude Code memory file. Use when the user wants to capture their writing DNA, build a personal style guide, or create a reusable memory profile from their existing content.
tashanzhishi
A workflow for turning YouTube, Bilibili, podcast, or article content into a readable Chinese note stored under a knowledge folder called 他山之石. It separates the source author’s facts and opinions from the note writer’s interpretation.
spool
Search your local Claude Code and Codex CLI session history.