itmegirish
12 verified skills
.claude/skills/civil-classification
# SKILL: civil-classification ## Purpose Complete civil suit classification taxonomy for Indian courts. Use when building/updating LKB entries, intake prompts, or cause_type lists. Every cause type in the LKB must map to exactly one group below. --- ## AXIS 1: CAUSE GROUPS (16 groups, 85 causes) ### 1. MONEY_AND_DEBT (15 causes) | # | cause_type | Description | |---|-----------|-------------| | 1 | money_recovery_loan | Loan recovery (hand loan, bank loan, promissory note) | | 2 | money_reco
development
.claude/skills/draft-benchmark
# SKILL: draft-benchmark Use when: user asks to run draft pipeline benchmarks, fill Draft_test.xlsx, compare with ChatGPT-5.4, or analyze draft errors. ## Instructions for Claude When this skill is invoked, follow these steps ONE BY ONE: ### Step 1: Run Pipeline on Each Scenario (Fill "Draft Agent" Column C) For each of the 10 scenarios in `docs/Draft_test.xlsx` column B: 1. Read the scenario text from the Excel file using openpyxl 2. Run the drafting pipeline: ```python import async
development
.claude/skills/legal-drafting-workflow
# SKILL: legal-drafting-workflow ## Purpose Execution guide for the drafting pipeline. v5.1 is running. v11.0 is target. ## Architecture Reference - CLAUDE.md: `.claude/CLAUDE.md` (v11.0 target architecture) - Skills: `/v9-architecture`, `/schema-builder`, `/lkb-enrichment`, `/prompt-builder` --- ## How to Run ### Run single draft ```bash agent_steer/Scripts/python.exe research/run_draft_live.py ``` ### Unit tests ```bash agent_steer/Scripts/python.exe -m pytest tests/drafting/ -v ``` ###
development
.claude/skills/section-drafter-prompt
# SKILL: draft-prompt ## Purpose Build and refine the draft prompt that produces a complete court-ready legal document in a single LLM call. v5.1 uses free-text drafting — the LLM outputs the entire document (not section-keyed JSON, not gap-fill). Exemplar-guided, LKB-informed. ## When to Use - Building or modifying `prompts/draft_prompt.py` - Debugging why draft quality is low - Tuning exemplars or context injection - Adapting prompt for a new cause type - Optimizing prompt token count ## Ar
development
.claude/skills/section-validator
# SKILL: section-validator ## Purpose Build and maintain the 4 deterministic verification gates (Stage 3). Gates run on the full draft text with zero LLM calls. They validate, auto-fix formatting, and flag issues for review. ## When to Use - Building or modifying any gate - Adding new entity extraction patterns - Debugging false positives / false negatives - Extending verified provisions coverage ## Architecture Context (v5.1 — what's running) 4 gates run sequentially on `draft.draft_artifac
development
.claude/skills/test-draft-pipeline
# SKILL: test-draft-pipeline ## Purpose Run the drafting pipeline, evaluate output quality, and verify all 4 gates + review work correctly. ## When to Use - After modifying any pipeline node, gate, or prompt - After creating or updating an exemplar or LKB entry - For regression testing across multiple scenarios - For debugging pipeline failures ## Test Runners ### Quick Test (single scenario) ```bash agent_steer/Scripts/python.exe research/run_draft_live.py ``` ### Unit Tests ```bash agent_
development
.claude/skills/schema-builder
# SKILL: schema-builder Use when: creating, reviewing, or modifying document schemas for the v11.0 pipeline. Covers all 12 civil document types. ## What Are Document Schemas Document schemas define the STRUCTURE of a legal document — section order, per-section instructions, filing rules. They are INDEPENDENT of cause type. Same schema works for ALL 92 cause types. One schema per document type. ## The 12 Civil Document Schemas | # | Document Type | CPC Reference | Filed By | Annexure Prefix
development
.claude/skills/v9-architecture
# SKILL: v9-architecture Use when: planning, building, or reviewing v11.0 architecture components (LKB 2-layer model, document schemas, structured prompt builder, gates, family migrations). ## v11.0 Architecture — Scalable Context-Driven Pipeline ### Core Principles 1. **Better context to LLM = better draft** — no complex engine needed 2. **Separate law from structure** — cause type (92) × document type (12) = 1,104 combinations 3. **Decide law before drafting, enforce law after drafting** #
development
.claude/skills/draft-reviewer
# SKILL: draft-reviewer ## Purpose Build and maintain the review node (Stage 4). Review validates draft quality and optionally applies inline fixes. Uses slim payload — only draft text + gate errors + user request. ## When to Use - Building or modifying the review node - Tuning the review prompt or skip conditions - Debugging review output quality - Changing review model assignment ## Architecture Context (v5.1 — what's running) Review runs after 4 deterministic gates. It receives a slim pay
development
.claude/skills/lkb-enrichment
# SKILL: lkb-enrichment Use when: adding Layer 2 data to LKB entries, verifying LKB accuracy, or fixing wrong legal data. ## What Is LKB Enrichment LKB entries currently have Layer 1 (legal knowledge). v11.0 needs Layer 2 (document components) added to each entry. This skill guides enrichment. ## LKB 2-Layer Model | Layer | Fields | Purpose | |-------|--------|---------| | Layer 1: Legal Knowledge | `primary_acts`, `limitation`, `facts_must_cover`, `mandatory_averments`, `defensive_points`,
testing
.claude/skills/prompt-builder
# SKILL: prompt-builder Use when: building, reviewing, or debugging the structured prompt that combines LKB + document schema for the draft LLM. ## What Is the Prompt Builder The prompt builder is the core of v11.0. It replaces the 400-line `_build_lkb_brief_context()` flat dump with a ~100-line structured prompt builder that combines: 1. **Document schema** → section order + per-section instructions 2. **LKB Layer 1** → legal knowledge (statutes, limitation, facts guidance) 3. **LKB Layer 2
development
.claude/skills/template-builder
# SKILL: exemplar-builder ## Purpose Create, validate, and maintain document schemas and LKB Layer 2 data for the v11.0 scalable drafting pipeline. **v11.0 approach:** No exemplar documents in prompts. Instead: LKB 2-layer data + document schema → structured prompt → LLM drafts. ## When to Use - Creating a new document schema (e.g., written_statement, appeal_memo) - Enriching LKB entries with Layer 2 data (available_reliefs, jurisdiction_basis) - Reviewing schema quality against CPC rules - A
development