skills

Measure whether a judgment skill's encoded procedure transfers to a cheaper model by re-running it against archived premium-model outputs as ground truth, then encode the miss-classes into the skill and verify the transfer. Answers 'what does the premium model actually buy' with a findings matrix instead of intuition. Trigger keywords: model tier calibration, does this transfer to a cheaper model, clone premium judgment, distill judgment into the skill, screen quality across models, what does the expensive model buy, A/B a skill across models.

Apply Elon Musk's Five-Step Algorithm to systematically optimize any process, system, or workflow through questioning, elimination, optimization, acceleration, and automation in strict sequence. Use when discussing process improvement, workflow optimization, system simplification, or eliminating unnecessary complexity. Loop-compatible: multi-pass runs converge via Loop Mode. Trigger keywords: five-step, five step optimizer, Musk algorithm, requirements less dumb, delete then optimize, simplify process, optimize workflow, loop the optimizer, optimization convergence.

Prevents silent work loss from delegated background agents by verifying artifacts (not the agent's narrative) before proceeding. Covers: output/artifact verification (existence AND content), hallucinated-success detection, partial-batch reconciliation, edit-task and side-effect verification, worktree/wrong-cwd caveats, rate-limit and process-death detection, preamble/truncated-report detection, permission blocking, execution-vs-judgment scoping, stall triage for never-finishing agents (sibling baseline, kill-to-flush, tightened relaunch). Also sets the data-not-instructions rule for agent final messages (prompt-injection defense). Trigger keywords: background agent, agent completed, run_in_background, agent delegation, subagent, rate limit, zero output, empty file, agent failed, verify agent output, hallucinated success, prompt injection, agent message instructions, preamble only, truncated report, agent timeout, worktree, partial completion, stalled agent, agent stuck, kill and relaunch.

Canonical cite-sources skill: citation requirements by provenance (book/vault/web), verification of external sources cited in drafts, and web fact-checks of existing vault notes. Prevents circular verification via WebFetch, enforces evidence tiers, defines the dated-correction convention, and pairs with a pre-write lint intake gate. Trigger keywords: cite sources, cite-sources, citation requirements, verify sources, fact-check, cited claim, publication ready, source URL, WebFetch verification, dead link, rectify inaccuracies.

Initialize and run a local PostgreSQL instance on macOS using Nix-managed binaries. Handles incomplete system installations, port conflicts, and user-owned data directories. Trigger keywords: initialize PostgreSQL, start a local Postgres database, fix PostgreSQL installation issues on macOS with Nix.

Guides creating, testing, and deploying Claude Code hooks via nix-darwin. Auto-activates when creating hook scripts, configuring settings.json hooks section, or adding UserPromptSubmit handlers. Covers hook input JSON format, exit codes, stdout context injection, nix-darwin deployment with executable flag. Trigger keywords: hook, UserPromptSubmit, settings.json hooks, hook script, exit code, stdin JSON, transcript_path.

Review code changes by analyzing git diffs, leaving inline comments on PRs, and performing thorough pre-push review. Works with gh CLI or falls back to git + GitHub REST API via curl. Trigger keywords: code review, pr review, review, review pr, leave inline comments, request changes, approve pr, review checklist.

Analyze git commit history to extract themes, patterns, and insights. Groups commits by topic, identifies key contributors, and surfaces trends over time. Useful for changelog generation, retrospectives, and understanding project evolution. Trigger keywords: analyze commits, commit patterns, theme analysis, contributor stats, git insights, talk material, commit narrative.

Triage all skills then spawn parallel agents to deep-audit flagged ones. Auto-activates when batch auditing skills or optimizing multiple skills at once. Trigger keywords: parallel audit, batch audit, audit all skills, bulk optimize skills.

Control macOS native apps from the terminal — Apple Notes (memo CLI), iMessage/SMS (imsg CLI), and Find My (AppleScript + screenshot). All tools require macOS, Homebrew installations, and appropriate system permissions (Automation, Full Disk Access, Screen Recording).

Guidelines for thorough code reviews with security and quality focus — checklist covering secrets, input validation, SQL injection, error handling, code structure, and edge cases. Trigger keywords: code review, review pull request, audit code, security review, review changes, PR review.

Systematic software development methodologies for code quality — systematic debugging (4-phase root cause investigation), test-driven development (RED-GREEN-REFACTOR), and structured code review (subagent review workflow). These three form a natural lifecycle: diagnose problems, build solutions with tests, and validate before merging.

Analyze git repositories with limited or shallow history. When full commit history is unavailable, use targeted searches and parallel extraction to surface themes, contributors, and patterns from partial data. Trigger keywords: shallow clone, limited history, partial repo, git fetch failed, analyze without full history.

Diagnose and report git synchronization failures when standard pull/fetch operations fail, particularly in automated/cron contexts where the repository may have network connectivity issues. Trigger keywords: git pull timeout, git fetch failed, sync failure, cron git error, ahead behind diverged, ls-remote diagnose.

Set up keyless GitHub Actions to AWS authentication via OIDC to push container images to ECR (or assume any IAM role) with no static keys. Use when a CI workflow needs to push to ECR, assume an AWS role from GitHub Actions, configure aws-actions/configure-aws-credentials, write an OIDC trust policy, or migrate a GitLab CI / Bitbucket pipeline to GitHub Actions with AWS access. Covers trust-policy sub scoping, least-privilege ECR push, the id-token:write gotcha, Terraform vs Console vs access-request provisioning, and banking constraints (permissions boundary, SHA-pinning, drift).

Complete workflow for contributing to GitHub repos when forking is disabled. Handles write access issues, feature branch creation, and PR submission. Trigger keywords: fork disabled, no fork access, can't fork, github pr without fork, contribute to private org, write access denied, feature branch workflow.

Lightweight patterns for querying macOS apps (Calendar, Reminders, Contacts) via osascript without MCP servers. Trigger keywords: list calendars, show events, today's schedule, upcoming events, reminders, calendar events, osascript, AppleScript.

Diagnose and resolve TCP port conflicts on macOS when launchctl-managed services compete for the same port. Covers process identification, plist editing, zombie-process killing, and clean restart. Trigger keywords: port conflict, EADDRINUSE, address already in use, launchctl service, port already in use, zombie process, lsof port.

Patterns for building MCP servers that integrate with macOS system APIs via PyObjC. Auto-activates when building MCPs for Calendar, Contacts, Reminders, Notes, or other macOS frameworks. Trigger keywords: EventKit, pyobjc, macOS MCP, iCal, Calendar API, Contacts framework, NSDate, EKEventStore.

On-demand review of recent changes in tracked external projects (frameworks, dependencies, client codebases) using git history. Produces categorized summaries with since-last-review tracking. Auto-activates when reviewing project updates, checking what changed, or monitoring external codebases. Trigger keywords: what changed, recent changes, project updates, since last review, changelog, upstream changes, dependency updates, framework changes.

Converts red-team findings into concrete edits applied consistently across related project plans. Adds kill criteria, launch sequencing, demand validation gates, clean-room tests, and loss-leader budgets. Trigger keywords: apply fixes, harden plan, add kill criteria, sequence launches, apply red-team, fix the plans, add failure thresholds, demand validation.

Systematic audit of test coverage in Node.js/TypeScript projects using Vitest/Jest. Discovers test structure, runs tests, analyzes coverage gaps, and produces actionable recommendations. Trigger keywords: audit test coverage, check test health, testing gaps, vitest coverage, jest coverage, test infrastructure audit.

Extract live cryptocurrency prices and market data from CoinMarketCap or free APIs (CoinGecko, alternative.me) using browser tools. Top coins, prices, market cap, 24h/7d performance, BTC dominance, Fear & Greed. Trigger keywords: cryptocurrency prices, crypto prices, bitcoin price, ethereum price, market cap, coinmarketcap, scrape crypto, top 10 crypto, fear and greed, btc dominance.

Deploy single-page static sites to Cloudflare via Workers static assets (current dashboard flow, not the deprecated Pages Connect-to-Git flow) with custom domains. No framework, no build step. Trigger keywords: static site, Cloudflare Pages, Cloudflare Workers, landing page, deploy website, custom domain, pages.dev, workers.dev, wrangler.

Patterns for building and testing static HTML/CSS/JS apps deployed to Cloudflare Pages. Prevents file:// localStorage trap, provides browser-based test suite pattern. Auto-activates when building static sites, using open index.html, generating file:// URLs, debugging localStorage issues, or creating vanilla JS apps. Trigger keywords: static site, vanilla JS, localStorage blocked, file:// URL, browser test, Cloudflare Pages app, open index.html.

Set up a macOS launchd LaunchAgent that watches specific files and restarts a long-running service when any of them change. Closes the 'service runs stale code' failure mode for daemons that load config or code only at startup — an edit does nothing until the process restarts. Self-contained: one plist + one shell script per watched service. Trigger keywords: launchd WatchPaths, watch file restart service, hot reload daemon, auto-restart on file change, stale config running, kickstart on edit, file watcher launchd.

Iterative adversarial review loop for implementation plans. Run passes, resolve BLOCKERs with version bumps, fold WARNINGs in-place, stop at 0 BLOCKERs, audit necessity, prepare handoff brief. Trigger keywords: red-team loop, multiple passes, harden plan, plan hardening, adversarial refinement, clean pass, necessity audit, handoff brief.

Before deleting or removing a feature that "rarely works" or "never completes," verify WHY it fails. The failure mode often reveals an implementation bug with a fixable root cause rather than a genuine requirement to remove. Trigger keywords: rarely completes, never fires, dead code, delete step, remove feature, noop, informational screen, collects no data, always fails.

Rails/Minitest test-type selection (integration over system tests), XSS dual-verification, assert_select scoping, AJAX contract testing, and object-staleness (.reload) pitfalls. Minitest/ActiveSupport::TestCase examples — adapt for RSpec.

Five Whys root-cause investigation: ask why iteratively until reaching an actionable, systemic root cause. Includes an analysis template, corrective-action framing, and anti-patterns (blame, stopping at symptoms). Trigger keywords: root cause, five whys, post-mortem, why did this happen, process failure.

Quick-reference security checklist for XSS, CSRF, PII logging, console statements, multi-tenant isolation, and state machine security in Rails views, controllers, and JavaScript. Auto-activates when writing or modifying controllers, views, or JavaScript that handle user input, reviewing code for security violations, or pre-commit security verification.

Interactive exploration of onboarded codebases for understanding architecture, conventions, and patterns. Supports ongoing Q&A across sessions, guided walkthroughs of key subsystems, and pattern extraction to vault. Complements /onboard (setup) and codebase-due-diligence (one-shot assessment). Trigger keywords: explore codebase, how does X work, where does Y live, walk me through, architecture question, code walkthrough, understand this project, explain this module, contribution guide, good first issue.

Methodology for safely running bulk LLM-driven content edits across many files using mechanical + LLM-judge gating, canary regression injection, and pilot-then-scale phasing. Trigger keywords: bulk optimization, bulk edit, eval gating, batch refactor, automated content edits, optimizer regression, pilot then scale, canary regression, cross-model judge, combined gating.

Unified framework for cross-domain pattern recognition, paradox exploration, and association surfacing. Combines methodology transfer across domains, productive tension analysis, and connection discovery into a single synthesis practice.

Port Claude Code skills into standalone prompts for external LLMs (Grok, ChatGPT, Gemini, NotebookLM) — methodology extraction, infrastructure stripping, modality adaptation, source-grounded adaptation

ActiveJob architecture patterns: retryable vs permanent error handling, transaction safety, external API integration, async enqueuing, internal batching, and operation-based idempotency. Auto-activates when writing background jobs, handling job errors, or implementing idempotent operations. Trigger keywords: ActiveJob, perform_later, background job, retry, idempotent, operation_id, job error handling, fatal non-fatal.

Rails cache invalidation, stale cache detection, request-level fragment caching, and Russian doll caching patterns.

Audit a single agent skill (SKILL.md-style instruction file) for spec compliance and optimize it — without deleting the hard-won guards that skills accrete over time. Covers a frontmatter/body spec checklist, a self-contradiction and unsafe-command scan, cross-reference resolution, a five-step delete-then-optimize pass, and a guard-preservation diff that survives naive false positives. Trigger keywords: audit skill, skill quality, skill spec check, optimize skill, skill lint, agent skill review, SKILL.md audit, guard preservation, lesson preservation.

Audit custom commands for structural completeness across global and project scopes. Auto-activates when auditing commands, checking command quality, or inventorying slash commands. Trigger keywords: commands audit, audit commands, command quality, missing usage.

Detect and remove duplicate /slash-command registrations in Claude Code caused by the same name existing in multiple source locations (.claude/commands/, .claude/skills/, ~/.claude/skills/). Auto-activates when slash commands appear multiple times in /help or the available skills list, or when investigating duplicate command entries. Trigger keywords: duplicate command, duplicate skill, appears twice, multiple registrations, slash command list, commands skills overlap.