houfu

Finds and recommends existing blog posts as internal backlinks for a post at the final draft stage. Use when a draft is nearly complete and needs internal links to related published content.

Comprehensive draft and post access, creating, editing and analysis. When Claude needs to work with the Ghost Admin API to access content published on alt-counsel.com as Houfu's partner.

Coordinates audience feedback on a draft or idea from 1-3 specialized reviewer agents (legal-tech blog, in-house lawyer, lawyer-coder personas), with an enforced 2-round cap and length-audit-first ordering. Use when the user wants audience feedback on a draft, idea, or question. The /feedback command is a shortcut that runs all 3 reviewers; this skill chooses reviewers by content type.

Develops a focused pitch (title plus 150-200 word paragraph) that defines the scope, thesis, and suggested tags for a new article or newsletter. Use at the start of drafting any new post, after the idea is clear enough (use brainstorming first if it is vague).

Research legal technology topics for blog posts on alt-counsel (Ang Hou Fu's blog). Use when the user asks to research topics, find sources, fact-check claims, gather statistics, or find expert opinions for blog content. Prioritizes Singapore/ASEAN perspectives and flags US/EU-centric information. Outputs research findings to research.md in the post folder with proper citations.

Interactive idea refinement using Socratic method to develop fully-formed designs when partner describes any blog or newsletter idea, before writing a pitch or articles

Suggests and validates tags from the canonical registry (tags.md) to prevent tag sprawl and keep naming consistent. Use when creating a new post, during pitch generation, before publishing, or when the user asks about tags.

Record discussions in the posts folder on an ongoing basis when requested to, or at the end of significant events, like brainstorming, generating a pitch etc

Process large volumes of bank statements (50+ PDFs) in batches with checkpointing and progress tracking. Orchestrates the standard ingestion skill across multiple batches for resumable processing.

Categorize financial transactions by analyzing merchant names, amounts, and context. Assigns spending categories with confidence scores and suggests pattern rules. Use when you need to categorize transactions or determine spending patterns.

Parse bank statement CSV files into structured transaction data. Extracts account information and transactions from CSV exports. Use when you need to process CSV bank statements.

Extract transactions from PDF or CSV bank statements in the staging folder, categorize them using Claude AI, insert into database, and archive processed files. Use when you need to ingest, process, or import new bank statements (PDF or CSV format).

Manage transaction memory to view patterns, add quick notes, search for merchants, and clean up outdated entries. Use when you need to check what's in memory, add new learnings, or maintain the memory file.

Parse bank statement PDF text into structured transaction data with account information and transactions in consistent JSON format. Works with any bank format. Use when you need to extract or parse transactions from PDF bank statements.

Interactively review flagged transactions to confirm or correct categorizations, and create rules based on user feedback.

Template analyzer for CoQuill (v2). Parses docx/HTML templates, extracts variables including conditionals and loops, merges config.yaml overrides, infers types, and generates a v2 manifest.yaml. Called by the coquill orchestrator — not triggered directly by the user.

Document renderer for CoQuill. Takes a template, variable values, and produces rendered documents (docx or html+pdf). Validates output for unfilled placeholders. Called by the coquill orchestrator — not triggered directly by the user.

Document assembly tool. Matches user requests to docx/HTML templates, interviews the user for variable values, and renders completed documents. Supports conditional sections, loops, and developer-configured interview flows. Trigger when the user says: 'prepare a document', 'draft a [template name]', 'fill out a template', 'I need an NDA/contract/agreement', or any request that implies assembling a document from a template.

Transcript generator for CoQuill. Reads an interview_log.json and manifest.yaml, then writes a human-readable transcript.md to the job folder. Called by the coquill orchestrator — not triggered directly by the user.

Deep-dive, multi-agent review of a single pull request in the target project. Invoke ONLY when the user explicitly runs /lq-maintainer:review-pr N (N = PR number) — skill invocation is namespaced by the plugin; there is no bare /review-pr. Never invoke proactively, never mid-conversation on your own judgment — a review skill firing unprompted is surprising with no upside. For batch or single-item triage without the deep dive, the user runs /lq-maintainer:triage instead.

Triage the target project's PRs and issues into recommended lanes and draft Triage Receipts. Use ONLY when the user explicitly runs /lq-maintainer:triage (batch digest across all open PRs and issues), /lq-maintainer:triage pr N (single PR), or /lq-maintainer:triage issue N (single issue). Skill invocation is namespaced by the plugin — there is no bare /triage. Never invoke proactively or mid-conversation without an explicit command.