Skip to content
Agentmart
Submit a tool

library-qa

Use when an attorney asks a question that should be answered only from the firm's vetted reference-library manuals in Torq, such as a statute, a playbook or a…

Skill by Torq-Legal-Labs

IndexedMIT
Write a review

Chalk Labs

Need help deploying library-qa?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/torq-legal-labs-library-qa

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when an attorney asks a question that should be answered only from the firm's vetted reference-library manuals in Torq, such as a statute, a playbook or a procedure guide, with a citation for every claim.

Use cases: testing, research

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 2, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/library-qa

What's inside

Skill name
library-qa
Description
Use when an attorney asks a question that should be answered only from the firm's vetted reference-library manuals in Torq, such as a statute, a playbook or a procedure guide, with a citation for every claim.
License
MIT
Instructions
17 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Ask The Library

Use this skill when an attorney asks a question that should be answered strictly from the firm’s vetted reference-library manuals (statutes, playbooks, procedure guides) rather than general model knowledge.

This is a closed-corpus, no-guessing surface. Never fill a gap with outside/model knowledge, even if you are confident the answer is correct in general — the value of this skill is that every answer is traceable to a specific manual.

Workflow:
1. Call semantic_search_docs (add space_ids only if the user named a specific manual/space) to retrieve candidate chunks. Never pass reference_library_only=false or scope on this surface. When the thread is pinned to the reference library the server forces the closed corpus and ignores those arguments, so a widened search is not available — never describe one to the user.
2. STOP with "not covered in the reference library" when the result reports notCovered=true, or when the top hit’s score < 0.615 (the calibrated not-covered similarity threshold). On a STOP, name the space(s) searched (spacesSearched) and do not proceed to step 3 — never answer from a below-threshold chunk.
3. Otherwise (covered=true), compose the answer using ONLY the retrieved chunk text. Attach a citation to every claim: manual title + section path + page number, drawn from each chunk’s citation object.
4. If a citation needs more surrounding context than the chunk carries, use kb_get_page_content (full page) or kb_get_page_version (a specific pinned revision) for the same page_id before finalizing the answer — never widen scope to a different page or manual to fill a gap.
5. If retrieved chunks conflict or only partially cover the question, say precisely which part is covered and which part is not, rather than blending in an assumption.
6. Never present a "not covered" answer as if it were a considered legal opinion, and never present a covered answer without its citations attached.

Torq tools this guide uses: `semantic_search_docs`, `kb_get_page_content`, `kb_get_page_version`. They come from the Torq connector. If they are not available, Torq is not connected or the member's Torq permissions do not include them; setup is at https://torqlegallabs.com/ai-assistants.

From Torq-Legal-Labs/torq-claude-plugin, MIT.

More from Torq-Legal-Labs/torq-claude-plugin

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 17, 2026 (18 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Torq-Legal-Labs

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get library-qa running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying library-qa?

CallGet help
Cite this page

library-qa skill. Agent Mart. https://agentmart.dev/skills/torq-legal-labs-library-qa (last updated 2026-10-24)