You are an adversarial reviewer. You receive an investigation summary and a set of constraints/decisions. Your job is to find blind spots, flawed assumptions, and missing considerations.
You are an impartial code reviewer. You receive ONLY the diff, the Constitution, and relevant spec ACs. You have NO implementation context — no plan, no task description, no "what we were trying to do.".
You receive a specific review finding and make a targeted fix. You do NOT redesign, refactor, or improve surrounding code. You fix exactly what the finding describes.
Runs after a tool call finishes for Bash(git worktree remove:*) tool calls, executing worktree-safety.py via python. From joshsmithxrm/power-platform-developer-suite.
Run manifest-driven PNG captures of TUI and extension surfaces for design audits. Emits a schema-conformant capture tree for consumption by a Claude design-audit session.
Create, triage, and manage GitHub issues per PPDS backlog conventions. Use when creating issues, grooming the backlog, reviewing what to work on next, or filing bugs.
Interactive debugging with systematic root-cause analysis. Use when encountering any bug, test failure, or unexpected behavior — before proposing fixes.
Triage and merge open dependabot PRs per docs/MERGE-POLICY.md — classify each PR (auto-merge / verify-then-merge / manual review), enable auto-merge for safe ones, run targeted test suites for risky ones, and surface anything needing human judgment. Use when there's a backlog of dependabot PRs, after a quiet period…
Brainstorm ideas into specs and plans through collaborative dialogue. Use when starting a new feature, exploring an idea, or designing a system — before any implementation.
Building or modifying VS Code webview panels and multi-surface panel features. Use when creating new panels, adding features to existing panels, designing panel architecture across Extension/TUI/MCP/CLI.
Mechanical pass/fail checks. No judgment, no opinions - compiler, linter, and tests either pass or they don't. Run before code reviews, after fix batches, and before PRs.
Create PR, wait for the configured automated review, triage every comment, and present a summary. Use when work is ready to ship — after gates, verify, QA, and review are complete.
QA — Three-Agent Blind Verification. Use when making code changes that need product-level validation — after implementation, after fix rounds, before PRs. Dispatches agents that test the running product without seeing source code.
Cut a PPDS release — CHANGELOG refresh, version bump, tag push sequence, CI monitoring, post-publish verification. Use when preparing a new prerelease or stable release across CLI, TUI, MCP, Extension, and NuGet libraries.
Structured multi-surface product validation — systematically test Extension, TUI, MCP, and CLI with parity comparison and architecture audit. Use before releases, after large features, or to kick the tires.