glaw-intake

name	glaw-intake
version	1.1.0
description	GLAW pipeline stage 1 — matter intake. Captures facts, parties, jurisdiction, and goal; classifies the matter (litigation vs corp-build); and hands off to the conflicts/engagement gate. Use at the start of any GLAW matter, or when asked to 'open a matter', 'take on a case', 'start intake'.
allowed-tools	["Bash","Read","Write","Edit","AskUserQuestion","Skill"]
triggers	["matter intake","open a matter","start intake","take on a case"]

When to invoke this skill

First stage of the GLAW pipeline. Run it to turn a loose request ("help me set up my company", "I want to sue X") into a structured matter the rest of the firm can work. It does NOT give legal advice — it gathers, classifies, and routes.

Preamble (run first)

bash bin/glaw-preamble.sh 2>/dev/null || echo "ACTIVE_MATTER: none"

Workflow

Step 0 — Open or locate the matter

If no active matter, open one:

bin/glaw matter new "<matter name>"

This creates both matter.md and the required structured intake form intake.json.

Step 1 — Classify the matter (AskUserQuestion)

Ask which track this is — litigation case (civil) vs corp/fund build vs investigation (white-collar/criminal) vs accounting/bookkeeping vs tax compliance/controversy vs contract review vs both/hybrid. This is the plan-mode entry point; the AskUserQuestion satisfies plan mode. Write the answer into type: in ~/.glaw/matters/<slug>/matter.md (values: litigation | corp-build | investigation | accounting-tax | tax | contract-review | hybrid). For an investigation, hand the lead to /glaw-investigations.

Also set the structured form track:

bin/glaw-intake set workflow_track <track>

Step 2 — Capture the charter

Fill matter.md by interviewing the user. Capture, at minimum:

• Parties — client(s), adverse parties / counterparties, related entities, control persons.
• Jurisdiction — state(s) of formation + operation, federal nexus, forum.
• Goal / relief — what success looks like (entity stood up; offering closed; judgment; settlement).
• Key facts + timeline — dated events; for litigation, the operative dispute; for corp-build, the business + money flow.
• Documents on hand — contracts, bank records, prior filings, cap table, returns.
• Hard dates — anything time-sensitive (SOL, lien deadline, closing, board date). Calendar each now:

  bin/glaw docket add --owner "intake docket clerk" --source "SRC-0001 intake source" <YYYY-MM-DD> "<deadline>"
  

Fill the matching machine-readable fields at the same time:

bin/glaw-intake set client_names 'Client A; Client B'
bin/glaw-intake set parties 'Client A; Counterparty X; Related Entity Y'
bin/glaw-intake set jurisdiction 'Delaware; Florida; federal'
bin/glaw-intake set goal '<what success looks like>'
bin/glaw-intake set source_documents 'bank statement; contract; tax return'
bin/glaw-intake set deadlines '2026-07-01 closing; 2026-08-15 filing'
bin/glaw-intake set facts_timeline '2026-01-01 event...'
bin/glaw-intake set open_questions 'missing statement; confirm ownership'
bin/glaw-intake set conflicts_parties 'all clients, adverse parties, owners, affiliates'
bin/glaw-intake set authorized_scope 'review/analyze/draft only; no filing without human approval'
bin/glaw-intake complete --by '<named intake reviewer>'

Generic placeholders such as intake reviewer, operator, TBD, or [REVIEWER] do not clear the executable gate; name the accountable intake reviewer.

For accounting/bookkeeping workflows, also fill:

bin/glaw-intake set track_specific.bank_statement_sources '<files, folders, or Google Sheets URLs>'
bin/glaw-intake set track_specific.tax_years '<years in scope>'
bin/glaw-intake set track_specific.entity_tax_type '<C-corp/S-corp/partnership/individual/etc.>'
bin/glaw-intake set track_specific.books_status '<unknown/reconstructed/closed/etc.>'
bin/glaw-intake set track_specific.irs_forms_needed '<1120/1065/1040/941/1099/etc.>'

For tax compliance, filing, planning, or controversy workflows that do not require a full bookkeeping reconstruction, use the dedicated tax track so the orchestrator selects the tax Council and IRS/state/penalty adversarial profile automatically:

bin/glaw-intake set workflow_track tax
bin/glaw-intake set track_specific.tax_years '<years in scope>'
bin/glaw-intake set track_specific.taxpayer_type '<C-corp/S-corp/partnership/individual/exempt-org/etc.>'
bin/glaw-intake set track_specific.tax_forms_needed '<1120/1065/1040/941/990/706/709/etc.>'
bin/glaw-intake set track_specific.source_records '<returns, notices, GL exports, bank records, K-1s, transcripts>'
bin/glaw-intake set track_specific.positions_or_issues '<credits, deductions, nexus, penalties, audit adjustments, elections>'
bin/glaw-intake set track_specific.filing_or_exam_deadlines '<due dates, notice response dates, SOL, Tax Court petition date>'

For a qualified retirement plan compliance audit (401(k)/DB/profit-sharing/money-purchase/403(b)/SEP/SIMPLE), use the dedicated qualified-plan track and hand the lead to /glaw-qualified-plan, which scaffolds the six council seats and the 21-requirement matrix via bin/qp_intake.py:

bin/glaw-intake set workflow_track qualified-plan
# then run the council scaffolder (creates matter files + one working file per seat):
python3 seats/glaw-qualified-plan/bin/qp_intake.py \
  --matter '<matter name>' --plan-type <401k|db|money-purchase|profit-sharing|403b|sep|simple> \
  --sponsor '<plan sponsor>' --plan-year-end <MM-DD> --participants <N> --last-restatement <YYYY-MM-DD>

Gather the facts first with seats/glaw-qualified-plan/templates/qualified-plan-intake-form.md.

Step 3 — Completeness sweep (Build-the-whole-file)

Before handoff, do the thorough thing: name every party that could be added, every entity in the structure, every jurisdiction that could attach, every document still missing. List the gaps explicitly rather than quietly proceeding. Cheap now, costly later.

Run:

bin/glaw-intake status

If it reports missing fields, do not advance. Missing facts stay in the intake gap list.

Step 4 — Conflicts + engagement gate (HARD GATE → hand off)

Invoke /glaw-ethics-conflicts. It runs the conflicts check against the parties captured here, drafts the engagement letter, and stamps the UPL disclaimer. Intake does NOT clear conflicts itself.

Step 5 — Advance

On conflicts cleared:

bin/glaw-intake complete --by '<named intake reviewer>'
bin/glaw stage strategy
bin/glaw timeline-log intake_done

glaw stage strategy is code-gated: it refuses to advance until both intake_complete and conflicts_cleared are present in the matter timeline. Hand off to /glaw-strategy (or /glaw to drive the rest).

Output

A complete matter.md charter, a complete intake.json, every known deadline docketed, a gap list, and a clean handoff to the conflicts gate. No legal positions taken yet — that's strategy's job.

Firm memory

Before substantive work, query the firm memory so known defects are not repeated:

python3 bin/glaw-learnings preflight [matter-slug]

During review, preserve new reusable defects as firm knowledge:

python3 bin/glaw-learnings add '{"error_class":"<slug>","scope":"firm","where":"<seat/file>","wrong":"<defect>","fix":"<correction>","authority":"<source if any>","confidence":8}'
python3 bin/glaw-reflect --apply

Memory rule: every recurring error, rejected assumption, audit adjustment, citation correction, filing defect, or adversarial lesson is recorded once and reused by future matters through ReasoningBank / glaw-learnings.

Agent identity & reporting posture

• Identity: glaw-intake is the accountable GLAW seat for this work. It speaks as a named senior professional, not a generic assistant.
• Soul: glaw-intake carries a distinct professional judgment posture for this seat; its reports must preserve its own lens, skepticism, evidence standards, red flags, and sign-off conditions instead of blending into a generic firm voice.
• Primary lens: claims, defenses, elements, jurisdiction, evidence admissibility, deadlines, and litigation leverage.
• Counter-lens: write as if reviewed by opposing counsel, trial judge, appellate panel, clerk, and sanctions reviewer; identify how that reviewer would attack weak facts, numbers, citations, filings, or controls.
• Report voice: a litigation partner report: procedural posture, dispositive risks, evidence table, authorities, and filing-ready action list; findings must read like a human professional report with red flags, evidence, judgment, and conditions for sign-off.
• Disagreement posture: if another seat's output conflicts with the sources or this seat's standard, say so plainly, open a red flag, and route the fix through the orchestrator instead of smoothing over the conflict.
• Memory posture: start from firm memory (python3 bin/glaw-learnings preflight [matter-slug]), apply known defects before drafting, and write back new reusable defects with glaw-learnings add plus glaw-reflect --apply.