Intent-alignment plugin for coding agents: requirement-analysis playbook, three-level intent funnel with file-in-the-loop persistence, and an optional red-blue adversarial review skill. DingTalk group channel and blocking gates live in the intent-gate-service sister MCP service.
Cross-agent coding harness: one curated skill canon (TDD, brainstorming, plans, reviews, shipping, design verification) plus mechanical safety gates and an autonomous loop via the yoke CLI.
Advisory SDLC discipline skills (brainstorming, source-check, RED-first) for AI coding agents. Advisory only — not enforced gates; upsells to the code-oz engine for enforced gates and different-model review.
Enforced SDLC gates + cross-family review for AI coding agents. Discovers and invokes the code-oz engine; the binary is the only writer of gates, events, and reviews.
Makes the agent write human-quality prose (no LLM tells) whenever it writes docs, READMEs, commits, comments, or replies. Ships the Alchemy ruleset as a skill.
Zero-dependency sound notifications for Claude Code. Plays audible alerts via PermissionRequest and Stop hooks — know when Claude needs approval or has finished, without watching the terminal. macOS / Linux / Windows.
A tiny phase gate for coding agents: your shell checks decide pass/fail (never the LLM), your workflow routes the agent (advance / retry / escalate / done). The agent drives, headsign holds the state.
Work a GitHub issue backlog to zero, one issue at a time: /plan-backlog proposes the backlog from a vision document, /implement-issue takes a single issue from plan to PR, and polako work supervises the whole queue unattended.