docs-steward
claudeDocs maintenance pipeline — builds canonical indexes of a repo, audits docs for drift, duplication, orphans, and onboarding gaps, then actively edits docs and opens a PR.
1 command · 18 agents
Analyze a sent-mail export to produce a data-driven written voice profile, then optionally generate a personalized ghostwriting skill.
Plugin voice-forge from marketplace claude-skills in schmimran/claude-skills. Commands read from the repo's manifest — 8 days ago.
/voice-forge:analyze-writing-voiceAnalyze a sent-mail export to produce a data-driven written voice profile. Guides export, auto-discovers formats, parses, computes stats, verifies quotes, and writes a findings doc.
/voice-forge:build-ghostwriting-skillTurn a voice-forge findings doc into a personalized ghostwriting skill for Claude Desktop. Requires an existing findings doc from /analyze-writing-voice. Confirms before starting.
voice-analystPhase 2 — Runs analyze_voice.py against the merged dataset, reads results.txt, and interprets aggregate voice statistics into a structured narrative for the findings doc.
voice-example-readerPhase 3 — Surfaces intentional writing examples from the dataset, extracts verbatim excerpts grouped by rhetorical category, and runs the mandatory verify_quotes.py gate before any quote proceeds.
voice-findings-writerPhase 4 — Assembles the final voice-findings markdown from verified statistics and quotes only, following the findings template. Never includes unverified content.
voice-parserPhase 1 — Scans the archive drop directory for .mbox/.olm/.pst files, runs the appropriate parser for each, merges all results into a single dataset, and proves output before reporting any number.
voice-skill-builderWorkflow 2 — Synthesizes a personalized ghostwriting SKILL.md from a voice-findings doc, using the ghostwriting skill template. Pushes verified examples to a references file.
Docs maintenance pipeline — builds canonical indexes of a repo, audits docs for drift, duplication, orphans, and onboarding gaps, then actively edits docs and opens a PR.
1 command · 18 agents
Daily bug-discovery sweep for Node.js, Swift, and Kotlin repos — runs build/audit + multi-agent code review, filters false positives, and files confirmed bugs as GitHub Issues for feature-creator to remediate.
1 command · 6 agents
Schmim Knews · Dead Air — turns News Blind Spots dashboard rows (or any topic) into Instagram carousels and reported deep-dive Reels, with a review page and an Airtable log.
3 commands · 3 agents
Feature and bug-fix pipeline — plans, reviews, and implements GitHub issues labeled `feature - ready for claude` or `bug - ready for claude`.
1 command · 5 agents
Multi-tool security audit for Node.js web apps and Supabase projects — files findings as GitHub Issues with deduplication and auto-close.
1 command · 5 agents
Supervisor-led full-stack mobile feature development — intake, master plan, parallel iOS/Android implementation, cross peer review, and reconciliation on one branch. No PR, no mobile build.
1 command · 2 agents
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.