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.