By eumemic
General-purpose development skills: an autonomous loop driver (/loop) + iteration skills (kaizen, bughunt, shovel-ready, kaikaku), ship pipeline, retrospectives, git workflow, investigation, and testing
One iteration of speculative bug-hunting — audit codebase against project invariants and risk patterns via parallel agents, form hypothesis, prove with failing test, fix at root cause, hand off to /ship, run /retro, merge, declare the outcome. Use when invoked as `/bughunt` (runs one iteration); use `/loop bughunt` to run it as an autonomous loop. Pass `--auto` to merge automatically once CI is green; default is to wait for the user to merge in the GitHub UI. Project-agnostic; reads risk dimensions from CLAUDE.md key-invariants and project memory.
Reference document shared by the build-iteration skills (/shovel-ready, /kaizen, /bughunt) for their common single-iteration machinery — branch, sync-with-master, /ship handoff, /retro pass, merge, and declaring the loop outcome. NOT user-invokable, and NOT the loop. The loop lives in `/loop`; this is one iteration's build cycle. Only load when a build-iteration skill directs you here.
This skill should be used when the user asks to "commit", "push", "create a PR", "open a pull request", "monitor CI", "fix CI", "fix the build", "request review", "merge", "merge the PR", or mentions any git-based development workflow operations.
This skill should be used when the user asks to "diagnose", "debug", "investigate", "why is this broken", "find the root cause", "figure out why", "look at the logs", "what's going on", or when the user corrects with "you failed to get to the bottom of it" or questions apparent success. Also use when multiple attempts at fixing the same issue haven't worked, or when a problem needs systematic investigation.
Continuous architectural simplification and correctness-by-construction — whole-system recon via parallel agents to find high-leverage opportunities to simplify-and-beautify a structure OR foreclose a whole defect-class by making an illegal state unrepresentable, at the module/layer/concept altitude; steelman each candidate against Chesterton's Fence, then file groomed design-issue(s) for every candidate that clears the gate — autonomously, labeled by validation (no build, no in-session approval gate). Use when invoked as `/kaikaku` (runs one iteration); use `/loop kaikaku` to run it as an autonomous loop. The architectural-altitude sibling of `/kaizen`. Project-agnostic; reads simplicity + beauty + correctness stance from CLAUDE.md and project memory.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
General-purpose development skills for Claude Code: an autonomous loop driver + iteration skills, ship pipeline, retrospectives, git workflow, investigation methodology, and testing.
| Skill | Trigger | Description |
|---|---|---|
| loop | "/loop <skill> [flags]" | The loop driver. Runs one iteration of an iteration skill, reads its declared LOOP-OUTCOME, then paces / escalates / re-arms. Owns cadence, idle escalation, and flag passthrough so the iteration skills stay pure. E.g. /loop kaizen --auto, /loop kaikaku, /loop shovel-ready --auto --drain. |
/loop)| Skill | Trigger | Description |
|---|---|---|
| shovel-ready | "/shovel-ready", "drain the shovel-ready queue" | One GitHub issue-label queue iteration: take the highest-ROI labeled issue (TDD, /ship, /retro, merge), or audit-and-refill when empty. |
| kaizen | "/kaizen", "accidental-complexity reduction" | One audit-and-simplify iteration: parallel-agent codebase audit → highest-ROI simplification → /ship → /retro → merge. |
| bughunt | "/bughunt", "speculative bug-hunting" | One bug-discovery iteration: risk-pattern audit → hypothesis → red test → root-cause fix → /ship → /retro → merge. |
| kaikaku | "/kaikaku", "architectural simplification", "correctness by construction" | One architectural iteration: whole-system recon → Chesterton's-Fence steelman → autonomously file groomed design-issue(s) for every candidate that clears the gate (multi-output, no approval step), labeled by validation. Spans both architectural cells — simplify-and-beautify and correctness-by-construction (foreclose a defect-class by making an illegal state unrepresentable). Grooms, doesn't build (the architectural-altitude sibling of kaizen). |
| build-cycle | (not user-invokable) | Reference: the single-iteration build skeleton (branch → /ship → /retro → merge → declare outcome) shared by shovel-ready / kaizen / bughunt. |
| Skill | Trigger | Description |
|---|---|---|
| ship | "/ship", invoked by build-cycle | Drive a working tree to a green PR: project checks → /simplify → code-review subagent → commit → push → CI monitor. Stops at green; caller handles merge. |
| retro | "/retro", "run a retrospective", invoked by build-cycle with --scope=iteration | Quality-gated retrospective: identify durable codifiable learnings; produce nothing on most invocations. |
| git-workflow | "commit", "push", "create a PR", "merge" | Full git workflow orchestration with safety checks. |
| investigate | "diagnose", "debug", "find the root cause" | Systematic investigation methodology for bugs and mysteries. |
| test | "test this feature", "verify this works" | Thorough testing with cleanup and go/no-go reporting. |
The loop is decomposed into two layers, so "do one unit of work" and "decide whether to do another" stay orthogonal:
shovel-ready, kaizen, bughunt, kaikaku) each do one unit of work and end by declaring LOOP-OUTCOME: <status>. They know nothing about looping./loop runs an iteration, reads that outcome, and owns all the looping — cadence, idle escalation, re-arming, flag passthrough. It never touches repo content.The build iteration skills share the build-cycle skeleton (branch → implement → /ship → /retro → merge → declare outcome); each owns only its target-acquisition logic and implementation discipline. kaikaku grooms a proposal instead of building, so it owns its whole iteration and skips build-cycle.
shovel-ready GitHub label (extrinsic; humans populate the queue).Each iteration applies a quality gate (Phase 1.5): "found something" never implies "worth shipping." A non-productive iteration (empty / gate_killed) is a healthy, expected outcome — /loop paces it with long-cadence idle, not a forced ship. The same principle applies to /retro — most retros produce no actionable findings, and that's the healthy default.
Install via the eumemic marketplace in Claude Code:
/plugins → eumemic → dev-skills
npx claudepluginhub eumemic/claude-plugins --plugin dev-skillsCheck Claude Max usage across multiple accounts
Autonomous AI-assisted development methodology that keeps specs and code in sync through intelligent iteration loops
End-to-end development loop: idea to reviewed PR with automated code review iterations
Clone websites by reverse-engineering them into specs using AI-powered browser exploration
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.