taskflow
claudeDrive one small task from idea to merged PR through ten fixed steps with approval gates: statement, stories, mockup, plan, TDD, review panel, cross-agent review, live pass, PR, summary.
1 skill · hooks
Parallel read-only review tracks over a diff: mechanical triage, one agent per lens, every must-fix finding checked against the code, persisted rounds, re-review until clean.
Plugin review-panel from marketplace vbp1-skills in vbp1/skills. Commands read from the repo's manifest — 12 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
break-it-panelSettle the review panel's adversarial predictions with real tests. Takes the concrete breaks the adversary track predicted — file, line, triggering input — writes a focused test per prediction, runs it, and keeps only what caught a real defect or covers a genuinely valuable edge case. Use after a re
review-panelRun a parallel panel of read-only review tracks over a scope of changes: mechanical triage, one agent per lens, every must-fix finding checked against the code before it is reported, a persisted round file, and re-review rounds until nothing must-fix or important survives. Use before committing, whe
review-adversaryPre-commit review track: actively tries to FALSIFY the changed code — concrete inputs or states that make it throw, return a wrong result, corrupt state, or violate an invariant. Read-only — it predicts breaks, it cannot execute the code or modify the working tree.
review-commentsPre-commit review track: comment accuracy against the code, and stale or misleading docs the change introduced or left. Read-only — holds no tool that can modify the working tree.
review-correctnessPre-commit review track: correctness bugs and adherence to CLAUDE.md conventions and lint rules. Read-only — holds no tool that can modify the working tree.
review-over-engineeringPre-commit review track (gate 8): over-engineering lens (ponytail-review, preloaded) — reinvented stdlib or existing helpers, unneeded dependencies, speculative abstractions, dead flexibility. Read-only — holds no tool that can modify the working tree.
review-silent-failuresPre-commit review track: swallowed errors, empty or over-broad catch blocks, inappropriate fallbacks. Read-only — holds no tool that can modify the working tree.
review-simplifyPre-commit review track (gate 8): PROPOSES simplifications that preserve behaviour. Advisory only — it holds no tool that can modify the working tree and never applies its own suggestions.
review-specPre-commit review track: checks the change against the task it was written for — requirements missing or partial, behaviour nobody asked for, requirements implemented wrongly. Every finding quotes the spec line. Read-only — holds no tool that can modify the working tree.
review-testsPre-commit review track: test coverage gaps in the change, and whether each bug fix ships a regression test. Read-only — holds no tool that can modify the working tree.
review-type-designPre-commit review track: encapsulation, invariants and nullability in the types the change introduces. Read-only — holds no tool that can modify the working tree.
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/review-agent-guard.py"
Drive one small task from idea to merged PR through ten fixed steps with approval gates: statement, stories, mockup, plan, TDD, review panel, cross-agent review, live pass, PR, summary.
1 skill · hooks
Restore muffled or low-quality recordings: stem separation, EQ correction, high-frequency recovery.
1 skill
Adversarial test pass: write tests that try to falsify a change, keep only the ones that catch real defects.
1 skill
Cross-agent review workflow: Codex implements, Claude Code reviews, findings loop back until resolved.
1 skill
Inspect web pages in a local headless Chromium, stealth on request: screenshots, client-side JS errors, failed requests, long-task timings. Claude Code variant.
1 skill
Inspect web pages in a local headless Chromium, stealth on request: screenshots, client-side JS errors, failed requests, long-task timings. Codex CLI variant.
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
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.
5 commands
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.