superpowers
claudecodexCore skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Spec-driven workflow cho Claude Code: intake → spec → plan → build (implement/QC/report). Duyệt bằng chat tự nhiên, hook nhắc bằng mã [TDQ:*] và kiểm bằng hiệu ứng thật, chế độ nhanh (express) / chuyên sâu (deep), working log bắt buộc. Một nguồn skills/ chạy trên Claude Code, Codex, OpenCode và Antigravity.
Plugin tdq-workflow from marketplace tdq-local in TDQUOC/TDQ-Workflow. Commands read from the repo's manifest — 4 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.
tdq-checkportableUse when a TDQ portable bundle has just been copied into a project, before running any other TDQ skill - verifies manifest.json checksums, Python/command/MCP requirements, and self-heals what is missing.
tdq-browser-researchDeep web research and SPA web scraping through Chrome DevTools MCP. Uses adaptive polling and exponential backoff timeouts to prevent missing delayed AJAX data.
tdq-browser-testE2E and visual UI/UX verification through Chrome DevTools MCP. Audits DOM structure, viewport responsiveness, element interactions, and console errors.
tdq-buildRun an approved TDQ plan end to end in one turn, QC it against the DoD, write the report, then ask about the commit. Use right after a deep-pipeline plan is approved.
tdq-check-statusRecover an unfinished TDQ request: read the disk directly, report, and continue after a single nod from the user. Use when an old session died, when switching machines, or when another agent just ran a phase for you.
tdq-conventionsShared TDQ workflow rules (one-turn protocol, state, approval, hook reminder codes, git, working log, research). Loaded by the other tdq-* skills; never invoked directly.
tdq-intakeOpen a new TDQ request - record the ask, pick the lane, init state, analyse + interview until nothing is vague. Use for EVERY new prompt, including questions, checks and small jobs.
tdq-leanHunt over-engineering and name what can be deleted - review a diff, audit the whole repo, or harvest `ponytail:` markers into a debt ledger. Use when code feels bigger than the request.
tdq-planTurn an approved spec into a checkbox plan, one test per task: STOP and wait for the user to approve the plan, then ask how to run it and build in the same turn. Use when a deep-pipeline spec has been approved.
tdq-setupInstall and prove every dependency the workflow searches with: 7 rungs from the agent-lsp binary to the import-root config and the graphify graph. Use when opening a request, when a search tool fails, or on a new machine.
tdq-specWrite the spec for a TDQ request in the user's document language, register it into state, present it then STOP and wait for approval; once approved, write the plan in the same turn. Use when the deep pipeline has finished analyze.
tdq-statusReport the current TDQ state (request, lane, phase, execution mode, who approved what) and the exact next step. Use when the user asks where the workflow stands.
tdq-build-resolverRapid-response subagent fixing compiler errors, build failures, syntax faults, and type check breaks with surgical, minimal diffs.
tdq-implementerImplements one assigned task of an approved TDQ plan in an isolated git worktree, red-green, and reports results as structured data.
tdq-qc-testerIndependently verifies an implemented TDQ plan against its Definition of Done - reruns tests, probes edge cases and logging, reports PASS/FAIL with evidence.
tdq-refactor-cleanerCode cleaner eliminating dead code, removing duplicates, and enforcing clean code invariants with zero functional regression and deletion audit logs.
tdq-reviewerReviews a TDQ spec or plan file for gaps, contradictions, missing tests, and over-engineering. Read-only - reports findings, never edits.
tdq-security-reviewerIndependent security auditor checking code changes for OWASP Top 10, hardcoded secrets, injection vulnerabilities, and least-privilege compliance. Read-only.
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/session_start.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/session_start.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/session_start.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/subagent_start.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/subagent_start.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/subagent_start.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/prompt_context.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/prompt_context.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/prompt_context.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/edit_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/edit_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/edit_gate.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/read_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/read_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/read_gate.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/bash_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/bash_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/bash_gate.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_gate.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/lsp_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/lsp_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/lsp_gate.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/ask_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/ask_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/ask_gate.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/search_observe.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/stop_gate.py" || py -3 "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/stop_gate.py" || python "${CLAUDE_PLUGIN_ROOT}/hooks/scripts/stop_gate.py"
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Scaffold, validate, and publish new Claude Code plugins with proper structure
3 skills · 1 command · 1 agent
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
Use Alibaba Open Code Review from local Codex.
2 skills
Claude Code skill for creating and presenting Slidev slidedecks
1 skill
Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.
12 skills · 1 MCP server · 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.