Hook
Runs when you submit a prompt, before the agent sees it, executing hook-launcher.mjs via node (2 commands). From sdsrss/claude-mem-lite.
Hook
Runs when you submit a prompt, before the agent sees it, executing hook-launcher.mjs via node (2 commands). From sdsrss/claude-mem-lite.
Plugin Claude Code
Plugin marketplace listing 1 plugin: gsd.
Plugin Claude Code
AI orchestration tool for Claude Code — GSD management shell + Superpowers quality core.
MCP server Claude CodeCodexCursor +2
MCP server "gsd" as configured in sdsrss/gsd-lite. Runs ${CLAUDE_PLUGIN_ROOT}/launcher.js with node.
Agent
Systematic debugging with root cause analysis.
Agent
Execute one task with TDD/self-review and return structured result.
Agent
Research domain ecosystem before planning.
Agent
Two-stage code review after executor completes.
Command
Run diagnostic checks on GSD-Lite installation and project health.
Command
Start project from requirements document or description text. Use when: (1) user provides a requirements doc, PRD, spec file, or design document, (2) user pastes a detailed feature description or task list, (3) user says '按这个需求做' or 'here\'s what I need built', (4) user has a clear written specification ready to plan…
Command
Resume project execution from saved state with workspace validation.
Command
Interactive project start — discuss requirements, research, plan, then auto-execute. Use when: (1) user wants to implement a new feature or complex functionality, (2) user wants to research/analyze a technology then build something, (3) user describes a multi-step task that needs planning and decomposition, (4) user…
Command
Display project progress overview derived from canonical state fields.
Command
Save current state and pause project execution.
Plugin Claude Code
Plugin marketplace listing 1 plugin: claudemd.
Plugin Claude Code
Personal AI-coding discipline harness — one developer's opinionated AI-CODING-SPEC v6.25, enforced via Claude Code hooks + shipped with the plugin.
Instructions file
Instructions for sdsrss/claudemd, covering code graph (repo-wide ast index) and claude-mem-lite — persistent memory.
Command
Read-only spec ↔ implementation coherence audit. Cross-references core ↔ extended §EXT refs, Sizing line accuracy (±20B), §0.1 headroom caps, MEMORY.md ↔ files bidirectional integrity, and §10 quick-check terms ↔ banned-vocab.patterns coverage. Findings graded CRITICAL/HIGH/MEDIUM/LOW. Borrowed from github/spec-kit's…
Command
Aggregate claudemd rule-hits over the last N days. Shows top banned patterns, hook deny counts, per-spec-section heatmap, and bypass-escape-hatch usage.
Command
R3 Step 2 — lesson-bypass detector. Joins memory-prompt-hint suggest events with subsequent transcript activity to compute cite-recall (applied / (applied + bypassed)) across recent sessions. Makes the §11 MEMORY.md read-the-file effectiveness observable.
Command
Clean up leftover claudemd-sync- / claudemd-memtags-hay- sentinels and claudemd-(mockgh|work). test sandbox dirs from $TMPDIR, stale tool-exhaust (session scratchpads, old fixtures) from /.claude/tmp per the spec §7-EXT retention window (mtime > TMPRETENTIONDAYS, default 7), and orphaned per-session sentinels from…
Command
Generate a thin, fact-based DESIGN.md from a UI project's real design-token sources and wire it into project CLAUDE.md. Use when (1) the user asks to configure design specs / 配置设计规范 / 设计规范 for agents, (2) UI sessions keep inventing colors or spacing instead of using the project's existing tokens, (3) a UI project…
Command
Run health checks on claudemd installation. Flags missing deps, spec drift, settings.json issues, hook drift, backup inventory, rule-usage health, MEMORY.md tag specificity, cross-layer memory maintenance (promote/repatriate/stale candidates). Supports --prune-backups=N.
Command
Bootstrap claudemd in the CURRENT Claude Code session (copy spec into /.claude/, install hook manifest, evict legacy entries). Use right after /plugin install claudemd@claudemd to skip the wait-for-next-session restart.