Compound V — a multi-model orchestrator for Claude Code, layered on Superpowers: a deterministic router (Opus for risky work, Sonnet for the mechanical), cross-model review (Claude + Codex), 3 scouts (code · docs · domain), semantic memory, and Epic/Marathon autonomous loop mode. The pill that turns one freelancer into a squad.
Command
Capture one genuine architecture decision as a thin, human-confirmed ADR under docs/superpowers/adr/NNNN-slug.md — decision-with-alternatives-and-consequences, references verified to exist, draft→confirm→commit, then FTS5-recallable via /v:remember. ADR only — no C4 diagrams, no SAD documents.
Command
Run Compound V Phase 1A — code archaeology — on a topic without running the other pre-flights. Useful when you want the technical audit but already have domain + library context.
Command
Re-run the collect + scope-gate + review tail of a Compound V run by run-id. Normalizes each job's output into results/ .json, re-runs the git-derived scope gate on every job, then the three-pass Review Gate (AC-gated). Use to re-check a dispatched run without re-dispatching workers.
Command
Open the Compound V observability dashboard — a present-only, read-only browser view over docs/superpowers/execution/ (runs, epics, per-job status, scope-gate verdicts, blocker ledger). emit writes a static self-contained HTML snapshot; serve runs an ephemeral, localhost-only, read-only live viewer. Observe in the…
Command
Execute a Compound-V-ready plan, manifest, or run-id via batched parallel multi-backend dispatch. Accepts a bare plan path (auto-materializes the manifest), or a manifest|run-id (dispatches directly). Runs partition-reviewer first; on PASS dispatches Task 0 serially then the parallel batches, enforces the scope gate…
Command
Drive an EPIC — chain several features into one autonomous, resumable, dependency-ordered build on a single branch. Each feature runs through the FULL v1.0 pipeline (recon-gated spec → 3 pre-flights → writing-plans + partition → manifest → dispatch → 3-pass review) in topological order, accumulating onto the current…
Command
Initialize Compound V in this project — detect backends and capabilities (Codex, Context7, required skills/agents), walk through any missing installs one at a time, pick a routing stance, and save project + user config.
Command
(Re)index docs/superpowers prose into the local V-memory cache so recall is current. Incremental by file hash; runs fully offline (FTS5, pure stdlib). Optionally enable the semantic lane with a one-time bootstrap. Run it after pulling new docs, or when /v:remember looks stale.
Command
Refresh the Compound V tier→model map — discover the concrete models each backend (claude, codex, antigravity, cursor, devin, opencode) currently offers, show them, let you assign deep/standard/light, and write the result into .claude/compound-v.json so intent-based routing survives model churn without touching any…
Command
Scan this repo and build a trusted, citation-verified knowledge base (docs/superpowers/architecture/) plus an AGENTS.md/CLAUDE.md bridge, behind a human approval gate; --refresh re-checks staleness.
Command
Materialize a Compound V execution manifest from a plan (or an accepted fast-path pre-eval). Applies the routing policy and writes manifest.yaml plus an initial state.json into docs/superpowers/execution/ / — without dispatching. Binds every pre-eval-backed run. Use this to inspect or edit the manifest before…
Command
Deep, two-axis, stack-agnostic code review of a pull/merge request or a local diff — review-only, never edits code. Runs the pr-review skill. Works on GitHub (gh), GitLab (glab), or a hostless local branch. Argument = PR/MR URL or number; empty = current branch vs base.
Command
Manage Compound V decision preferences — your OWN dated past-reasoning memory (memory + challenge, never a clone). Present-only, observe-in-output / control-via-CLI. stats shows override/disagreement rates and demoted/expired patterns; distill regenerates the in-repo (secret+PII-scrubbed) preferences.md; show prints…
Command
Recall relevant past Compound V context — prior decisions, failures, routing lessons, specs/plans — from docs/superpowers via V-memory (FTS5, plus semantic embeddings when bootstrapped). It surfaces EVIDENCE for planning and review, never an authority over routing. Use before planning a feature, before reviewing a…
Command
Resume an interrupted Compound V orchestrator run by run-id. Reconciles state.json against git reality (git-wins tie-break) and re-dispatches only the incomplete jobs (pending/failed/blocked) via Engine A, then continues collect → review → merge.
Command
Run an independent cross-model (Codex) adversarial review of a Compound V plan/manifest before dispatch, then arbitrate the findings. Codex advises; the orchestrator decides.
Command
Render the state of a Compound V orchestrator run — pipeline phase plus a per-job status table — by reading state.json from the run directory. Optional run-id argument; without one, list runs and pick the most recent.