Plugin Claude Code
Plugin marketplace listing 1 plugin: claude-apd.
Plugin Claude Code
Plugin marketplace listing 1 plugin: claude-apd.
Plugin Claude Code
Agent Pipeline Development — enforced multi-agent pipelines with mechanical guardrails.
Agent Claude Code
Builder agent for the backend layer — API endpoints, services, repositories.
Agent Claude Code
Reviews code changes for bugs, security issues, edge cases, and cross-layer mismatches.
Agent Claude Code
Builder agent for the frontend layer — React components, pages, hooks.
Hook
Runs when a session starts on startup, running bash. From zstevovich/claude-apd.
Hook
Runs before the agent uses a tool for Agent, Bash, Bash, Bash, Read, Glob, Grep, Read, Write, Edit, Bash, Write, Edit, Bash, SendMessage, Write, Edit, Write, Edit, Write and Edit tool calls, running bash (12 commands). From zstevovich/claude-apd.
Hook
Runs after a tool call finishes for Bash tool calls, running bash. From zstevovich/claude-apd.
Hook
Runs when a subagent starts, running bash. From zstevovich/claude-apd.
Hook
Runs when a subagent finishes, running bash. From zstevovich/claude-apd.
Hook
Runs before the context is compacted, running bash. From zstevovich/claude-apd.
Hook
Runs after the context is compacted, running bash. From zstevovich/claude-apd.
Hook
Runs on the PermissionDenied event, running bash. From zstevovich/claude-apd.
MCP server Claude CodeCodexCursor +2
Model Context Protocol SDK. Runs locally from the mcp Python package.
Skill Claude CodeCodex
Use when verifying that APD is correctly configured on Codex in the current project — qualitative deep audit of agents under .apd/agents/, AGENTS.md, MCP server registration, .codex/hooks.json, and pipeline health. Goes deeper than apd:apddoctor. Triggers on "audit APD", "review setup", "is APD configured", "verify…
Skill Claude CodeCodex
Use BEFORE writing the APD spec-card.md and calling apd:apdadvancepipeline('spec', ...) on Codex whenever the task is vague, broad, ambiguous, or has multiple reasonable interpretations. Ask ONE question at a time, present 2-3 approaches when there are real choices, converge on a design the user explicitly approves.…
Skill Claude CodeCodex
MANDATORY before re-dispatching the builder after any APD verifier block or critical reviewer finding on Codex. Use on any bug, test failure, build failure, error, crash, regression, or unexpected behavior. Four-phase systematic debugging — root cause investigation, pattern analysis, hypothesis + controlled test, fix…
Skill Claude CodeCodex
MANDATORY after every successful APD pipeline commit on Codex — when apd:apdpipelinestate shows nextstep='commit' and a commit was made. Verify tests from a clean state, show the pipeline report, present four options (push, push+PR, keep local, discard), execute only after the user picks. Triggers on "after commit"…
Skill Claude CodeCodex
Use when the project has GitHub Projects configured (GITHUBPROJECTSURL in AGENTS.md) on Codex and APD pipeline tasks need to sync with the board. Creates issues for specs, moves cards through columns (todo → in-progress → review → done), closes on commit. Triggers on "GitHub Projects", "issue", "board", "card"…
Skill Claude CodeCodex
Use when the project has a Miro board configured (MIROBOARDURL in AGENTS.md) on Codex and the APD pipeline dashboard needs updating. Visualizes current pipeline status, completed tasks, recent metrics, and active spec cards. Triggers on "Miro", "dashboard", "board update", "visualize pipeline", "pipeline view"…
Skill Claude CodeCodex
MANDATORY before EVERY pipeline task on Codex — load BEFORE writing spec-card.md and calling apd:apdadvancepipeline('spec', ...), on every new task, no exceptions. It is the APD operating manual, NOT a brainstorm — "the task is already clear" is not a reason to skip it. Carries the current gate contract - phase map…
Skill Claude CodeCodex
MANDATORY during the APD builder phase on Codex (between apd:apdadvancepipeline('spec', ...) and apd:apdadvancepipeline('builder')) whenever implementing a feature, fixing a bug, or writing new code. Write a FAILING test first, watch it fail, write minimal code to pass, refactor while green. Every file write goes…
Agent
Second-pass reviewer — hunts for regressions and missed edge cases.
Agent
Writes or updates tests — writes inside tests/ only.