claude

pev

PEV (Plan-Execute-Validate) agent workflow with axiom-graph integration — /pev-cycle (full) and /pev-instance (slim)

0 on GitHubby ddpoev2.0.1MITupdated 14 days agoproductivity
/

Plugin pev from marketplace axiom-graph in ddpoe/axiom-graph. 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.

What it installs

Skills (12)

  • axiom-annotations-markers

    Reference for axiom-annotations decorator and step marker syntax. Use when adding or modifying @workflow, @task, Step, or AutoStep markers in Python code.

  • pev-architect

    Behavioral instructions for the PEV Architect planning phase — scope decision, codebase exploration, and writing a Shape Up-style pitch to the cycle manifest

  • pev-audit-annotations

    PEV Audit — backlog cleanup of annotation rule violations (duplicate step_num, sequence gaps, undecorated AutoStep targets), prose drift in docstrings/level_1/level_2 summaries, and coverage-gap candidates. Sequential single-fixer dispatch with batched human gate before any inline mechanical fix lan

  • pev-audit-consumer-docs

    PEV Audit — pre-release sweep of user-facing consumer prose. Three-phase orchestration (Discovery → Verification → Consolidate+Propose) with sequential subagent dispatch. Discovery walks dev sources (cycles / features / ADRs) for additions; verification stress-tests each consumer doc's claims via th

  • pev-audit-dev-docs

    PEV Audit — backlog cleanup of graph-staleness drift in dev docs. Sequential feature-shard dispatch with whole-slice plans and pass-staged execute (ghost → backlog). Use when accumulated drift in `docs/features/**`, `docs/adrs/**`, `docs/plans/**`, etc. has grown past what an in-cycle Auditor can ad

  • pev-auditor

    Behavioral instructions for the PEV Auditor validation phase — reads Builder's change-set, reviews stale nodes, updates docs, writes the Impact Report to the cycle's audit::impact-report

  • pev-builder

    Behavioral instructions for the PEV Builder execution phase — reads its Architect task, implements with TDD, writes its builder doc, returns a control envelope

  • pev-cycle

    PEV orchestrator — Plan-Execute-Validate workflow. Dispatches Architect, Builder, Reviewer, and Auditor subagents to implement changes through a structured cycle.

  • pev-doc-reviewer

    Behavioral instructions for the PEV Doc Reviewer phase — drift scanner for documentation the Auditor's graph-based workflow doesn't cover

  • pev-instance

    Slim PEV mode for small, well-scoped tasks, one after another, each in its own worktree — mini-pitch + human gate (the checkin is cloned there) + implement + update affected docs + commit on the branch, then an independent Reviewer and Doc Reviewer in parallel and the shared merge step. The one agen

  • pev-reviewer

    Behavioral instructions for the PEV Reviewer phase — reviews Builder code against Architect pitch for spec compliance, functionality preservation, and code quality

  • pev-spike

    Smoke-test all PEV hooks — creates a worktree, dispatches the spike agent with tiny budget limits, and reports pass/fail results for every hook.

Subagents (10)

  • pev-architect

    PEV Architect — reads codebase via axiom-graph tools, writes Shape Up-style pitch to the cycle manifest

  • pev-audit-annotations-fixer

    PEV Audit Annotations-Fixer — classifies annotation rule violations, prose drift, and coverage-gap findings; applies mechanical fixes inline via Edit/Write and drafts spawn-request payloads for risky/judgmental fixes

  • pev-audit-consumer-discovery

    PEV Audit Consumer-Discovery — walks dev-source trees (cycles, features, or ADRs by mode parameter) looking for absences (capabilities or user-impacting decisions not yet reflected in consumer docs). One agent file, three modes selected at dispatch.

  • pev-audit-consumer-verifier

    PEV Audit Consumer-Verifier — reads one consumer doc (or a small batch) and stress-tests every anchored claim against three layered staleness signals (LINKED_STALE, git-diff slice, capability cross-reference). Issues per-claim verdicts (still-true / now-wrong / needs-review / out-of-scope).

  • pev-audit-dev-shard

    PEV Audit Dev-Shard — plans and executes graph-staleness drift cleanup for one feature shard's slice of dev docs

  • pev-auditor

    PEV Auditor — reviews Builder's changes, updates docs, marks nodes clean, writes Impact Report

  • pev-builder

    PEV Builder — implements the Architect's pitch using TDD in an isolated worktree

  • pev-doc-reviewer

    PEV Doc Reviewer — reviews Auditor's doc changes against templates and implementation

  • pev-reviewer

    PEV Reviewer — reviews Builder's code against Architect pitch (spec compliance, functionality preservation, code quality)

  • pev-spike

    PEV Spike — smoke-tests all PEV hooks with tiny budget limits

Hooks (10)

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-worktree-scope.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-doc-scope-md.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-bash-scope.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-worktree-rm.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-docjson-guard.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-doc-scope.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-axiom-graph-scope.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-tool-gate.sh"

  • PostToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-tool-counter.sh"

  • SubagentStop

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pev-subagent-stop.sh"

View source on GitHub

More from ddpoe/axiom-graph

All 2 plugins →

Graduated plugin-hook test agents — isolates hook resolution behavior (CLAUDE_PLUGIN_ROOT vs CLAUDE_PROJECT_DIR, frontmatter vs hooks.json) with canary-file reporting and hook-input-JSON dumping

2 skills · 5 agents · hooks

0

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

caveman

claude

Talk like caveman. Cut filler. Keep technical details explicit.

22 skills · 1 command · 3 agents · hooks

JuliusBrussee/caveman 119,462

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Taste Skill

claudecodex

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

Leonxlnx/taste-skill 114,391

career-ops

claudecodex

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

career-ops-hq/career-ops 82,253

Maintain pev?

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.

claude & codex plugins2 plugins