claude

aiharness

Agent harness: hook-enforced implementation pipeline over git worktrees

1 on GitHubby Jeromev0.6.1MITupdated 24 days agodevelopment
/

Plugin aiharness from marketplace aiharness in marmelab/AIHarness. Commands read from the repo's manifest — 23 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use and it starts local MCP servers. The exact commands are listed below; read them before installing.

What it installs

Skills (15)

  • adr-writing

    Architecture Decision Record format and rules for the developer agent. Load when the implementation introduces a structural decision worth remembering 6 months later — new pattern, new dependency, deliberate departure from convention, non-obvious schema choice. Skip for naming and file-layout micro-

  • e2e-conventions

    When to write e2e tests, where to put them, and how to verify them. Apply to any task touching UI, filters, forms, or interactions.

  • grill-me

    Scope a #harness request before dispatching the orchestrator. Invoke on EVERY #harness request, in the main thread. Step 0 decides whether questions are needed and exits in one line when the scope is already precise, so the decision is always made and always visible.

  • plan-grill

    Question what the planner had to invent, at the harness plan gate. Load in the MAIN thread once the planner has written its tickets and before the coordinator asks for approval. Never inside a subagent, never before a plan exists.

  • playwright-testing

    Playwright E2E testing patterns — web-first assertions, user-visible locators, network interception, fixtures, authentication, and parallel execution. Use when building or reviewing E2E tests with Playwright, when setting up browser testing for a web app, or when migrating from Cypress or Selenium.

  • ponytail-audit

    Whole-repo audit for over-engineering. Like ponytail-review, but scans the entire codebase instead of a diff: a ranked list of what to delete, simplify, or replace with stdlib/native equivalents. Use when the user says "audit this codebase", "audit for over-engineering", "what can I delete from this

  • ponytail-debt

    Harvest every `ponytail:` comment in the codebase into a debt ledger, so the deliberate shortcuts and deferrals ponytail leaves behind get tracked instead of rotting into "later means never". Use when the user says "ponytail debt", "/ponytail-debt", "what did ponytail defer", "list the shortcuts", "

  • ponytail-help

    Quick-reference card for all ponytail modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /ponytail-help, "ponytail help", "what ponytail commands", "how do I use ponytail".

  • ponytail-review

    Code review focused exclusively on over-engineering. Finds what to delete: reinvented standard library, unneeded dependencies, speculative abstractions, dead flexibility. One line per finding: location, what to cut, what replaces it. Use when the user says "review for over-engineering", "what can we

  • ponytail

    Forces the laziest solution that actually works, simplest, shortest, most minimal. Channels a senior dev who has seen everything: question whether the task needs to exist at all (YAGNI), reach for the standard library before custom code, native platform features before dependencies, one line before

  • pr-description

    Writes a short, concise PR description following a fixed template, to avoid the "sprawling" descriptions the team systematically rewrites by hand. Invoke before opening a PR.

  • resolving-rollback-conflicts

    Replay a list of git-revert commits against the current base branch and resolve the conflicts, when the chat-service's automatic rollback hit a merge conflict it couldn't resolve. Load this when your developer dispatch asks you to resolve a rollback conflict (it carries BASE_BRANCH, FAILED_COMMIT, C

  • setup-interview

    Domain-by-domain interview to produce $CLAUDE_PROJECT_DIR/docs/project-context.json. Invoked once by the orchestrator; the orchestrator then conducts all turns directly using Read/Write/Edit — no agent dispatching.

  • worktree-detection

    How to find a Git worktree by task number without hardcoding paths. Use in hooks or scripts that need to locate a specific task's worktree.

  • writing-migrations

    Generate Supabase SQL migrations at deploy time from the session branch diff. Load this when your developer dispatch asks you to generate the deploy-time migration (it points you at the shared <base>/simple worktree). This is NOT a feature ticket — the workflow below replaces the normal ticket rules

Slash commands (10)

  • /aiharness:harness-diff

    Show what a harness session changed (the net diff of its work)

  • /aiharness:harness-revert

    Revert a harness session's changes and clean up its branches/worktrees

  • /aiharness:harness-review

    Review a diff with the harness reviewer (code quality, security, over-engineering) without changing anything

  • /aiharness:harness-target

    Point this harness session's fork base + promotion target at a branch (worktree workflows where $CLAUDE_PROJECT_DIR stays on main)

  • /aiharness:harness

    Run a change through the agent harness, asking for the level and the gate instead of making you remember them

  • /aiharness:ponytail-audit

    Audit the whole repo for over-engineering, what can be deleted

  • /aiharness:ponytail-debt

    Harvest ponytail: comments into a tracked debt ledger

  • /aiharness:ponytail-help

    Quick reference for ponytail levels, skills, and commands

  • /aiharness:ponytail-review

    Review changes for over-engineering, what can be deleted

  • /aiharness:ponytail

    Switch ponytail intensity level (lite/full/ultra/off)

Subagents (7)

  • developer

    Implementation agent. Spawned by the orchestrator (foreground) per ticket. Reads the ticket, plans, implements + commits in its git worktree, then emits an output contract line so the orchestrator can dispatch reviewers. Validation runs via SubagentStop hooks; the merger handles the merge.

  • documentator

    Two modes. Mode 1 — captures recurring patterns into runtime artifacts under $CLAUDE_CONFIG_DIR/local/, indexed in the patterns ledger. Mode 2 — at session end, appends business-knowledge bullets to $CLAUDE_PROJECT_DIR/MEMORY.md. Always orchestrator-triggered.

  • merger

    Local merge agent (no team, single-shot). Dispatch contexts — (1) per-task Stage A merge of a feature branch into the session branch, (2) SIMPLE / MIGRATION flow (Stage A then promotion in one shot, on the simple branch), (3) promotion-only (Stage B, session branch → base branch under flock), (4) RO

  • orchestrator

    Implementation orchestrator for code-change requests via the agent harness. The main thread dispatches this agent for any feature / bugfix / code-change request; it classifies (SIMPLE vs COMPLEX, plus SETUP / MEMORY / ROLLBACK-CONFLICT / RECOVERY), plans, dispatches developer/reviewer/merger waves i

  • planner

    Product task planner. Use at the start of any code-change request and at the end of FULL_SETUP (SETUP_MODE=true). Decomposes natural-language product needs into atomic, ordered, actionable tickets with best-guess file paths.

  • quality-reviewer

    Combined code quality, security, and QA review agent, the sole reviewer in a COMPLEX wave (code + security review AND runtime/integration validation), single-shot in the SIMPLE flow (every SIMPLE change is reviewed, at the tier the harness computed from its diff), and single-shot in `migration-revie

  • test-writer

    Optional test-only agent, off by default. Dispatched by the orchestrator (foreground, on the developer's worktree, before review) ONLY when a ticket carries "separate_test_writer": true. Writes or strengthens tests; never touches application code. Validation runs on its stop, same as the developer.

MCP servers (1)

  • playwright

    runs: bash ${CLAUDE_PLUGIN_ROOT}/scripts/playwright-mcp.sh

Hooks (23)

  • SessionStart

    ${CLAUDE_PLUGIN_ROOT}/hooks/session-bootstrap.mjs

  • SessionStart

    ${CLAUDE_PLUGIN_ROOT}/hooks/ensure-playwright-mcp.mjs

  • SessionStart

    ${CLAUDE_PLUGIN_ROOT}/hooks/link-session-workspace.mjs

  • SessionEnd

    ${CLAUDE_PLUGIN_ROOT}/hooks/cleanup-session.mjs

  • PreToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/pre-tool-bash.mjs

  • PreToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/restrict-documentator-write.mjs

  • PreToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/block-migration-writes.mjs

  • PreToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/pre-tool-agent.mjs

  • PreToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/require-workflow-agent-type.mjs

  • PostToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/format-on-write.mjs

  • PostToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/check-typescript-shortcuts.mjs

  • PostToolUse

    ${CLAUDE_PLUGIN_ROOT}/hooks/check-view-column-order.mjs

  • PostCompact

    ${CLAUDE_PLUGIN_ROOT}/hooks/reinject-invariants.mjs

  • Stop

    ${CLAUDE_PLUGIN_ROOT}/hooks/turn-complete.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/cleanup-worktree.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/record-review-verdict.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/record-developer-done.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/record-smoke-evidence.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/e2e-on-feature-review.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/validate-on-stop.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/completion-invariant.mjs

  • SubagentStop

    ${CLAUDE_PLUGIN_ROOT}/hooks/render-status.mjs

  • UserPromptSubmit

    ${CLAUDE_PLUGIN_ROOT}/hooks/remind-harness-persona.mjs

View source on GitHub

More from marmelab/AIHarness

All 2 plugins →

Per-session report of where minutes, tokens and dollars went, for any Claude Code session

3 commands

1

Similar plugins

superpowers

claudecodex

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

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

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

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

slidev

claude

Claude Code skill for creating and presenting Slidev slidedecks

1 skill

slidevjs/slidev 49,452

Maintain aiharness?

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