build-firm-brain
Use when a firm owner, or a member granted Brain authoring, wants to build, resume or deepen the firm's Firm Brain in Torq through its guided interview…
Skill by Torq-Legal-Labs
Chalk Labs
Need help deploying build-firm-brain?
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.
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-build-firm-brain
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.
About
Reviews
Write a reviewNo 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. Checked Oct 25, 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/build-firm-brain
What's inside
- Skill name
build-firm-brain- Description
- Use when a firm owner, or a member granted Brain authoring, wants to build, resume or deepen the firm's Firm Brain in Torq through its guided interview, section by section.
- License
- MIT
- Instructions
- 27 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Build The Firm Brain Use this skill when an owner (or a granted contributor) wants to build or deepen their Firm Brain by chatting. This is a GUIDED, SERVER-DRIVEN interview: the server owns the cursor, the required-section set, and the completion decision. Your job is to converse each section to a good answer and confirm it — never to decide when the Brain is "done". Pick the scope (the `kind` argument) up front and keep it fixed for the whole run: - onboarding: the owner's first pass over every section. - cascade: a contributor answering only the sections assigned to them. - gap_fill: a re-launchable deepening pass over the sections the Brain is currently missing (thin or skipped). Loop: 1. Call resume_brain_interview with the kind. Read currentSectionKey (where to resume), the per-section status + any proposed draft, and isComplete. If isComplete is already true, tell the user the required set is satisfied and stop — do not re-open resolved sections. 2. Work the section at currentSectionKey. Call answer_brain_section (kind + the user's message). Pass message:null on the very first turn of an unopened interview so the coach speaks the section's opening prompt. Relay the coach's message events to the user and gather their answer. 3. Keep looping answer_brain_section for the same section until an event with type "proposed" returns a section body. Do not fabricate the body yourself — use exactly what "proposed" carries (the user may refine it in another turn first). 4. Read the proposed body back to the user in plain language and get explicit confirmation. Then call confirm_brain_section with confirm:true and that body to write it. If the user wants to skip a non-required section, call confirm_brain_section with confirm:false. 5. confirm_brain_section returns the next currentSectionKey and the server-computed isComplete. Trust those: move to the returned currentSectionKey and repeat from step 2. When isComplete is true, stop — the required set is satisfied. Determinism and safety: 6. Completion is the server's required-section check, never your judgment. A section that scores thin is re-entered by the server through the gap set on a later gap_fill run — do not force it or mark the Brain complete to move on. 7. There is one interview per (firm, user): a second start resumes the in-flight run rather than restarting it, so always resume before answering. 8. A confirm outside the acting user's allowed sections (cascade/gap_fill scope) is rejected by the server — respect the scope you chose and confirm only sections the state offers. While a section is open: 9. Your job is to elicit the owner's answer and record it with answer_brain_section, not to fetch data. Do not call Salesforce, QuickBooks, ads, phone or any other connected-source tool to fill in, check or enrich their answer: what they tell you is the answer. If they ask an unrelated question, answer it, then bring the conversation back to the open section. Torq tools this guide uses: `resume_brain_interview`, `answer_brain_section`, `confirm_brain_section`. 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
- 0.565score
gameplan-design
Torq-Legal-Labs
Use when the member asks to design, build, edit, reorder, translate or preview a client's Gameplan in Torq, or asks what a save can change, which compliance…
SkillAFreshness: AClaude Code +9 - 0.565score
kpi-scorecard-advisor
Torq-Legal-Labs
Use when an owner or department head wants to build, prune or fix a team's KPI scorecard in Torq: choosing the few numbers that matter, setting targets from…
SkillAFreshness: AClaude Code +9 - 0.565score
training-request
Torq-Legal-Labs
Use when an employee wants to file a training request, flag a process issue or propose an improvement in Torq through its guided interview.
SkillAFreshness: AClaude Code +9 - 0.559score
firm-data-querying
Torq-Legal-Labs
Use when the member asks a question about the firm's Torq data that no specific Torq tool answers, such as counts, totals, trends or rankings across leads…
SkillAFreshness: AClaude Code +9 - 0.559score
marketing-local-services
Torq-Legal-Labs
Use when the member asks about the firm's Google Local Services Ads in Torq: the LSA scorecard, leads that are candidates for a rating such as spam, the LSA…
SkillAFreshness: AClaude Code +9 - 0.559score
memory-and-provenance
Torq-Legal-Labs
Use when the member asks what the firm's past Torq Brain conversations said about something, where a remembered fact came from, or why an earlier answer cited…
SkillAFreshness: AClaude Code +9 - 0.559score
owned-media-grounding
Torq-Legal-Labs
Use when the member asks what the firm has said publicly about a topic, how the firm frames it, or wants copy grounded in the firm's own published articles…
SkillAFreshness: AClaude Code +9 - 0.559score
reporting-and-run-history
Torq-Legal-Labs
Use when the member asks about saved reports and report recipes in Torq, whether a Torq engine is running or failing, an engine's recent run history, or the…
SkillAFreshness: AClaude Code +9 - 0.559score
case-operations
Torq-Legal-Labs
Use when the member asks about a case in Torq: its status, upcoming dates and appointments, open tasks, where it sits in a pipeline, or the files attached to…
SkillAFreshness: AClaude Code +9 - 0.559score
intake-coaching
Torq-Legal-Labs
Use when a manager asks about intake performance in Torq: what good intake looks like, the team's weakest skills, practice scenarios, or assigning scenarios…
SkillAFreshness: AClaude Code +9 - 0.559score
case-strategy
Torq-Legal-Labs
Use when an attorney wants defense strategy, tactics or investigative questions for a specific matter, grounded in and cited to the firm's vetted…
SkillAFreshness: AClaude Code +9 - 0.559score
library-qa
Torq-Legal-Labs
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…
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 17, 2026 (14 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Torq-Legal-Labs
- 0.565score
kpi-scorecard-advisor
Torq-Legal-Labs
Use when an owner or department head wants to build, prune or fix a team's KPI scorecard in Torq: choosing the few numbers that matter, setting targets from…
SkillAFreshness: AClaude Code +9 - 0.565score
gameplan-design
Torq-Legal-Labs
Use when the member asks to design, build, edit, reorder, translate or preview a client's Gameplan in Torq, or asks what a save can change, which compliance…
SkillAFreshness: AClaude Code +9 - 0.565score
training-request
Torq-Legal-Labs
Use when an employee wants to file a training request, flag a process issue or propose an improvement in Torq through its guided interview.
SkillAFreshness: AClaude Code +9 - 0.559score
finance-and-vendor-spend
Torq-Legal-Labs
Use when an owner or billing member asks about the firm's finances in Torq: the current financial snapshot, trends over time, vendor spend and the largest…
SkillAFreshness: AClaude Code +9 - 0.559score
firm-data-querying
Torq-Legal-Labs
Use when the member asks a question about the firm's Torq data that no specific Torq tool answers, such as counts, totals, trends or rankings across leads…
SkillAFreshness: AClaude Code +9 - 0.559score
officer-intelligence
Torq-Legal-Labs
Use when the member asks about an arresting officer in Torq: finding one by name, badge or department, the officer's profile and history, or documents linked…
SkillAFreshness: AClaude Code +9
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 build-firm-brain 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.
Cite this page
build-firm-brain skill. Agent Mart. https://agentmart.dev/skills/torq-legal-labs-build-firm-brain (last updated 2026-10-24)