ansonphong/meta-dev - Page 2

meta-overlord

Run meta-overlord in Codex. Trigger: meta-overlord.

meta-planner

Run meta-planner in Codex. Trigger: meta-planner.

meta-probe

Run meta-probe in Codex. Trigger: meta-probe.

meta-repair

Run meta-repair in Codex. Trigger: meta-repair.

meta-review-batch

Run meta-review-batch in Codex. Trigger: meta-review-batch.

meta-review-design

Run meta-review-design in Codex. Trigger: meta-review-design.

meta-security

Run meta-security in Codex. Trigger: meta-security.

meta-sweep

Run meta-sweep in Codex. Trigger: meta-sweep.

meta-ux

Run meta-ux in Codex. Trigger: meta-ux.

meta-version

Run meta-version in Codex. Trigger: meta-version.

meta-visual-critique

Run meta-visual-critique in Codex. Trigger: meta-visual-critique.

opus-execute

Run opus-execute in Codex. Trigger: opus-execute.

plan

Write and save a self-contained implementation plan for a fresh agent. Use for plan requests, medium changes, or implementation planning before code.

runbook

Coordinate related plans through a campaign runbook.

sonnet-execute

Run sonnet-execute in Codex. Trigger: sonnet-execute.

diagnose

Diagnose a focused failure and preserve evidence.

harden

Stress-test a plan before execution.

review

Review a scoped change through the shared protocol.

changelog-engine

Batch engineering changelog. Append entries during work, cut on demand or auto-threshold, draft release post, pair with version bump.

design-eval

Design quality rubric scoring — evaluates implementation against design doc on 4 dimensions

dod-contract

Generate definition-of-done contracts from task descriptions. Output structured acceptance criteria, test plan, rollback strategy, and blast radius.

headless-worker

Patterns for headless Claude execution via claude -p --output-format json. Tool allowlist tiers, env inheritance, model override for Sonnet routing.

plan-validation

Judgment-based plan quality checks — cross-reference integrity, dependency ordering, contract completeness, terminology consistency

repair-loop

Attempt auto-fix with 3-attempt cap, smallest-fix-first strategy, failure dossier to inbox. Use when verify step fails.

sniff-test

Grug-brain sniff test for plans AND code — detect over-engineering, complexity demons, vague tasks, and bad practices in plans/specs (primary) or code (secondary). Report-only, never edits. Invoked as /sniff-test. By default sniffs the subject under…

version-manager

Multi-repo, context-aware versioning. Declarative config, semver/calver/custom strategies, follows cascades, atomic version_files updates.

waterfall-tracking

Stand up and maintain a visible stage-level task list for the /meta-dev 6-stage waterfall. Use in autopilot/walk runs so the user can walk away and watch Brainstorm→Design→Plan→Harden→Execute→Review progress live. Stage granularity only — never mirrors…