reflection
claudeReflection loop: mine learnings into a diary, and observe→refine→search reusable task rules.
4 skills
End-to-end AI-agent task delivery: requirements, solution, plan, review, execute, and ship features through a structured plan-to-merge workflow.
Plugin planning-workflow from marketplace mistakenot-skills in mistakenot/skills. Commands read from the repo's manifest — 20 days ago.
address-feedbackWorks through all open PR review threads by fixing code, replying to reviewers, and resolving threads. Use when 'address feedback', 'fix PR comments', 'resolve review threads', or after a code review has left feedback on the PR.
code-reviewPerforms a structured code review of changes with severity labels covering correctness, security, performance, error handling, testing, and architecture. Use when 'code review', 'review this PR', 'review changes', or when a structured review of implementation changes is needed.
commit-taskVerifies completeness of all planning docs and commits them to main. Use when 'commit the task', 'finalize the task', 'commit task docs', or after all planning docs have been reviewed and approved. Not applicable when implementation has already started or for committing code changes.
complete-taskFinalizes a feature branch and merges it to main. Addresses remaining PR feedback, runs tests, writes feedback.md, merges the PR, exits the worktree, and verifies the merge. Use when 'complete task', 'finalize and merge', 'ship this task', or when the feature branch is ready to ship.
delegate-taskDispatches /execute-task for a planned task to a fresh background worker (herdr), freeing this session. Use when 'delegate task', 'send to executor', 'hand off this task'.
delegateHands a freeform prompt to a fresh background coding agent (herdr) — a worker in its own worktree, or a planner on main. Use when 'delegate this', 'send to an executor', 'run this in background', 'spin up a planner'. Not for planned tasks with docs (use delegate-task).
execute-taskAutonomously implements a planned task end-to-end using the coordinator-subagent pattern. Reads task docs, creates a worktree, dispatches subagents per phase, tracks progress, and opens a PR. Use when 'execute task', 'implement task', or after planning docs have been committed.
new-epicPlans an epic.html: user-facing direction, guard rails, and a breakdown into deployable, sequenced tasks. Use when 'new epic', 'create epic', 'plan an epic'. Not for a single task (use new-task).
new-planWrites the Plan tab and backfills AC traceability for a task. Use when 'write a plan', 'new plan', 'create the plan', 'plan the task', or after solution is approved. Not applicable when solution doesn't exist yet.
new-solutionGathers context, writes Verification and Solution tabs for a task. Use when 'write a solution', 'design the solution', 'new solution', 'explore approaches', or after requirements have been approved. Not applicable when requirements don't exist yet.
new-task-quickPlans a task end-to-end in one unattended pass: Requirements, Verification, Solution and Plan tabs, pauses only for open questions, then Codex-reviews the answered doc. Use when 'new task quick', 'quick task', 'plan end-to-end'. Not for stage-by-stage review (use new-task).
new-taskCreates a plan.html with Requirements tab for a new task. Use when 'create a task', 'new task', 'start a task', 'write requirements', or when a user describes a feature/fix to plan. Not applicable when executing an existing task, reviewing existing docs, or planning every stage unattended (use new-t
request-claude-reviewRuns a Claude Code review of task planning docs or any files (code, markdown, config), then resolves its comments. Use when 'request claude review', 'claude review task 042', 'have claude review <file>'.
request-codex-reviewRuns a Codex review of task planning docs or any files (code, markdown, config), then resolves its comments. Use when 'request codex review', 'codex review task 042', 'have codex review <file>'.
request-council-reviewRuns a parallel Claude, Codex and Grok council review of task planning docs or any files, merges de-duplicated comments, then resolves them. Use when 'council review', 'review with all three'. Not for one reviewer (use request-codex-review).
request-grok-reviewRuns a Grok CLI review of task planning docs or any files (code, markdown, config), then resolves its comments. Use when 'request grok review', 'grok review task 042', 'have grok review <file>'.
resolve-commentsResolves review comment threads — inline in planning docs/markdown/HTML or in .review.md sidecars — by fixing, rejecting, or escalating. Use when 'resolve comments', 'address comments', 'fix review comments', or after a review. Not for PR feedback (use address-feedback).
review-taskReviews task planning docs or any files (code, markdown, config) and leaves structured comment threads. Use when 'review task docs', 'review the plan', 'review task 042', 'review <file> and leave comments'. Not for PR/diff review (use code-review).
status-reportReports what every background worker is doing (herdr), flags unhealthy ones, and reaps finished ones. Use when 'status report', 'check executors', 'executor status', 'how are tasks going'.
task-feedback-analyserExtracts recurring patterns from completed task feedback into generalizable workflow rules. Scans feedback.md files and review threads, clusters by theme, and drafts rules with a strict 3-example minimum. Use when 'analyse feedback', 'extract rules', 'find patterns in feedback', or when reviewing hi
Reflection loop: mine learnings into a diary, and observe→refine→search reusable task rules.
4 skills
Build, run, validate, and manage A/B evals for skills in this repo: replay real tasks, compare skill versions on cost and quality, and validate the eval before trusting it.
2 skills
Structured ideation: generate high-impact feature ideas and run synthetic user-research simulations to decide what to build next.
2 skills
Design end-to-end assurance and testing strategies for autonomous agent-built software.
1 skill
Consult a council of independent coding-agent models (claude, codex, gemini, opencode, grok) for diverse, unled opinions on a question; strategies, members and prompt framings compose like an SDK.
1 skill
Customer discovery: mine real customer conversations for demand signals, jobs-to-be-done, and unmet needs.
1 skill
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.