bcanfield

Write or refresh a "Tech debt operations" section in the project's AGENTS.md so the team shares one source of truth for debt-ops disciplines. Run ONLY when the user explicitly asks to set up, install, or initialize debt-ops disciplines — never auto-invoke. Idempotent; only the managed section changes, other sections are untouched.

Register a deferred decision in the debt registry. Trigger by judgment, not a marker scan, whenever a future reader would ask "why this way?": an unmade decision, stub, loosened type, bypassed check, swallowed error, a default picked "for now", or a TODO/FIXME/HACK/XXX marker. Trigger immediately whenever you defer work, or when the user invokes /debt-ops:add. Over-register freely; the developer drops with "drop A", "drop A,C", or "drop all".

Write or refresh the `## Tech debt operations` section in CLAUDE.md so a team shares one source of truth for debt-ops disciplines and cached quality commands. Idempotent. Only the managed section changes; other sections are untouched. Invoked explicitly via /debt-ops:init (solo users get the same content from the SessionStart inject).

Print a debt-ops health summary from the metrics log, covering registration rate, hook feedback action rate, ADR creation, and AI-authored share. Use when the user asks for "debt-ops metrics", "debt health", "registry stats", or invokes /debt-ops:metrics. Read-only, never writes the log.

Audit the debt registry, rank survivors by churn × Fowler quadrant, surface a top-N list, then walk paydown on user follow-up. Use when the user asks to review debt, see what to pay down, work through entries, or invokes /debt-ops:review. Stale entries drop with `drop A,B,C`.

Register a deferred decision in the tech-debt registry. Trigger by judgment, not a marker scan, whenever a future reader would ask "why this way?": an unmade decision, stub, loosened type, bypassed check, swallowed error, a default picked "for now", or a TODO/FIXME/HACK/XXX marker. Trigger immediately whenever you defer work, or when the user asks to log/register tech debt. Over-register freely; the developer drops with "drop A", "drop A,C", or "drop all".

Write or refresh a "Tech debt operations" section in the project's AGENTS.md so the team shares one source of truth for debt-ops disciplines. Run ONLY when the user explicitly asks to set up, install, or initialize debt-ops disciplines — never auto-invoke. Idempotent; only the managed section changes, other sections are untouched.

Print a debt-ops health summary from the metrics log, covering registration rate, feedback action rate, ADR creation, and AI-authored share. Use when the user asks for "debt-ops metrics", "debt health", "registry stats", or a tech-debt health summary. Read-only, never writes the log.

Create, update, audit, and maintain AGENTS.md, CLAUDE.md, .cursorrules, GEMINI.md, copilot-instructions, or similar coding-agent instruction files, using evidence-based best practices. This covers setting up or bootstrapping such a file for a repo or service ("set up a CLAUDE.md", "bootstrap agent docs", "init"); cleaning up, trimming, shortening, or fixing a bloated file the agent seems to ignore; advice on what content belongs in these files; deciding between one root file vs. per-package files in a monorepo; and consolidating drifting per-tool files (Cursor, Claude Code, Codex, Copilot) into one canonical file with symlinks or imports. Also use when the user never names a file but wants coding agents to know repo-specific commands, conventions, or gotchas — they say "document this for the agent", "make sure Claude knows about X", "add this to the project instructions", the agent keeps guessing commands wrong or repeating a mistake, or a change added commands, env vars, or conventions agents must know. Hand

Write new prose or rewrite existing text so it doesn't read as AI-generated. Use whenever producing content a human will read as writing — blog posts, emails, READMEs, LinkedIn posts, docs, announcements, marketing copy — and always when the user says "humanize", "make this sound less like AI", "this sounds robotic", "rewrite this in my voice", or asks you to edit a draft for tone. Apply it even when the user doesn't mention AI: if the output is prose meant to sound like a person wrote it, this skill applies.

Write and run TypeScript scripts with Bun instead of plain Node.js or JavaScript — getting type safety, built-in tooling, and faster execution without any build step or extra config. Use this skill whenever the agent needs to create, run, or manage scripts and CLI tools. Prefer Bun over node, ts-node, tsx, npm, npx, jest, or vitest for a faster, more maintainable workflow. Also trigger for bun-specific keywords: bun, bunx, bun test, bun install, bun add, bun run, Bun.file, Bun.serve, Bun.$, bun:sqlite, bun:test.