stonematt

Read the Obsidian vault and summarize current context: active projects, recent reflections, and priorities from the last 7 days. Use when the user says "build my context", "what am I working on", "catch me up", "summarize my week", "current context", or asks for a status overview of their vault.

Run the user's daily vault harvest cycle: YouTube transcripts → voice captures → daily enrichment → phone calls → inbox processing. Each sub-skill is idempotent, so re-runs are safe. Use when the user says "do my daily stuff", "run my dailies", "daily cycle", "morning run", "run the daily", or similar.

Identify patterns coalescing across the vault — scattered thoughts clustering into potential projects, essays, or products. Use when the user says "what's emerging", "emerge", "find patterns", "what's clustering", "what ideas are forming", "spot trends in my notes", or asks about nascent themes ready to become something bigger.

Extract phone call transcripts from Apple Notes and save them to the Obsidian vault with link-on-first-mentions processing. Batch processes all unprocessed transcripts, tracks history to avoid reprocessing. Use this skill whenever the user says "process call transcripts", "get my calls", "import call recordings", "phone transcripts", "icloud calls", "call transcript", "process my phone calls", "get transcripts from notes", or mentions Apple Notes call recordings or phone transcript import.

Process and file notes from the Obsidian vault inbox (0.inbox/) using PARA method filing rules. Use this skill whenever the user says "process inbox", "file inbox", "clean up inbox", "process new notes", "what's in my inbox", "triage inbox", or mentions filing/organizing notes from the inbox. Also trigger when the user references 0.inbox/ and wants to take action on the notes there.

Process voice captures from iOS Siri dictation into the Obsidian daily note. Reads timestamped entries from an iCloud Drive capture file, scores transcription confidence, corrects Siri-mangled proper nouns using vault vocabulary, inserts cleaned entries into the daily note, and archives the raw content. Use this skill whenever the user says "process my captures", "voice captures", "get my voice notes", "import my random thoughts", "process random thoughts", "check my captures", "what did I dictate", "run obsidian capture", or mentions the Siri capture file, dictation processing, or voice note filing.

Document AI agent skills and hooks in the Obsidian vault with concept and implementation notes. Use when the user says "document this skill", "document this hook", "write up the skill", "add skill docs", "document skill to vault", "document hook to vault", or asks to create vault documentation for a skill or hook they've built or are building.

Enrich daily notes by harvesting activity from git repos, Apple Photos, and other sources. Automatically catches up missed days using a 7-day lookback window and per-host run history. Use when the user says "enrich daily", "enrich today", "harvest my day", "what did I do today", "fill in my daily", "daily roundup", or asks to populate their daily note with real activity data. Also trigger for "enrich yesterday", "catch up on yesterday", or any request to backfill daily notes with activity data.

Reconcile Claude Code settings.local.json (personal, gitignored) against settings.json (repo-tracked, shared): diff the two, promote stable rules to the shared file, drop duplicates, and flag secrets/machine-specific paths to keep local. Use when the user says "promote settings", "review settings", "clean up settings", "sync settings", "promote permissions", or asks about moving local settings to repo-tracked settings. This skill owns the local-to-shared promotion diff — a niche neither built-in covers. NOT for: editing settings.json in place, moving a permission to user scope, or adding env vars/hooks (use the update-config skill); auto-generating an allowlist from transcript noise to cut permission prompts (use /fewer-permission-prompts, which writes new rules to project settings.json but does not reconcile your existing local vs repo files).

Merge a pull request, clean up the branch, and optionally promote to production. Use this skill whenever the user wants to merge a PR — including /stone-merge, "merge it", "merge the PR", "land it", "merge and release", "merge to prod", "ship to production", or any confirmation after checks pass. Also trigger when the user says "merge after checks", "merge when green", or refers to merging + deploying in the same breath. This is the counterpart to /stone-commit — commit gets code ready, merge lands it and cleans up.

Create commits following Conventional Commits format using heredoc (not command substitution). Use this skill whenever the user wants to commit code changes — including /stone-commit, "commit this", "commit and push", "yes commit it", or any confirmation of a prior offer to commit (e.g., "yes do it", "go ahead", "yeah commit that"). Also trigger for "create a commit", "ship it", or when the user agrees to commit after completing a task. If the conversation involves creating a git commit in any form, use this skill. This skill overrides default commit behavior — always prefer it over built-in commit instructions.

Generate a dev journal entry for today's work — pulling signal from git commits AND the non-git context that matters in 6 months (vendor integrations, stack decisions, infra events, service signups, dead-ends worth remembering). Trigger when the user says /stone-journal, "journal entry", "write up today's work", "log today", or similar. Trigger even when the day had no commits but real work happened (accounts created, services configured, DNS changed, architectural decisions made).

Show a summary of dev journal entries. Trigger when user says /stone-journal-status, "journal status", "show journal", "journal entries", "what's in the journal", "list journal", or asks about journal contents. Use this instead of /stone-journal when the user wants to review existing entries rather than create new ones.

A self-review checklist for any drafted prose to catch AI-generated tells before a human sees it. Use this skill whenever you've written or are about to deliver written content (emails, posts, memos, docs, marketing copy, blog drafts, slack messages, READMEs) and want to make sure it doesn't read like AI wrote it. Trigger on phrases like "sniff test this", "smell test", "does this sound like AI", "make this sound human", "review my draft for AI tells", or whenever the user references the sniff test before publishing or sending. Other skills (e.g. make-screed, lithos-email) may invoke this skill as a quality gate.

Generate an executive-level client summary of product changes since a given date. Use when the user says /stone-client-report, "summarize for the client", "executive summary", "session report", "client update", "what did we deliver since [date]", or wants to roll up dev work into business-value terms for stakeholder communication. Also trigger when the user mentions preparing for a client meeting, writing a deliverables recap, or reconciling work against an SOW.

Run the MCP server sniff test against Claude Desktop. Swaps the Claude Desktop config to the local dev server, gives the user a test prompt, evaluates returned results, and restores the production config. Use when the user says "sniff test", "test in desktop", "desktop test", "test prompts in Claude Desktop", or wants to verify MCP tools/prompts work in Claude Desktop before publishing.