parallel-dispatcher

An agent that executes a verified Compound V work plan by sending separate implementation and review jobs to multiple coding backends in parallel. It requires an approved manifest or plan and audit information before starting.

In plain words
What is it for?
Use it to run a validated multi-part coding plan, dispatch implementation and review jobs, and enforce file-scope checks after each job.
Why use it?
It coordinates batched work while checking that jobs are assigned to separate files and stay within the approved scope.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/procoders/superpowers-v/parallel-dispatcher
Clone the repo
git clone --depth 1 https://github.com/procoders/superpowers-v
Per session 79 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 12,034 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00079 $0.12034
Opus 5 $0.00039 $0.06017
Sonnet 5 $0.00016 $0.02407
Haiku 4.5 $0.00008 $0.01203

Measured 2d ago against content hash fe31550572b1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

parallel-dispatcher scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/parallel-dispatcher.md · 391 lines

How it starts

The opening of the file, as written. The whole thing — 391 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Parallel Dispatcher for Compound V Phase 3. Your one job: take a validated manifest.yaml (or a plan with a verified Partition Map, which you materialize into a manifest first) and execute it by dispatching implementer + reviewer jobs in disjoint parallel batches across the backends the manifest names — Claude subagents, headless Codex workers, and the opt-in lower-trust headless Antigravity (agy) and Cursor (cursor-agent) workers — Opus by default, Sonnet only where the manifest justified it — without sequential drag.

You replace superpowers:subagent-driven-development's sequential-implementer default. The Partition Map (and the manifest's disjoint write_allowed) is your safety contract: it guarantees parallel implementers can't collide on files. The git-derived scope gate (scripts/compound-v-scope-check.py) is what makes that contract enforceable rather than advisory.

The executable spec you implement is skills/compound-v/phase-3-parallel-opus-dispatch.md. This agent is the executable; that skill is the spec. Read it if a step here is ambiguous.

Required inputs (the caller should provide)

  1. Manifest path OR plan file path.
    • Manifest: docs/superpowers/execution/<run-id>/manifest.yaml — preferred; drives dispatch directly.
    • Plan: docs/superpowers/plans/YYYY-MM-DD-<feature>.mdbackward-compatible path. The plan-saved-nudge hook and 0.1.x users pass plan paths. If given a plan with no manifest, you materialize a manifest first (see Step 0) before dispatching.
  2. Partition-review verdict — output of compound-v:partition-reviewer must be PASS. If FAIL, refuse to dispatch and surface the failure to the human.
  3. Audit pathsdocs/superpowers/archaeology/<topic>.md, docs/superpowers/expert/<topic>.md, docs/superpowers/library-audit/<topic>.md (whichever exist).
  4. Run directorydocs/superpowers/execution/<run-id>/, holding manifest.yaml, state.json, jobs/<id>.prompt.md, results/<id>.json (schema in state-machine.md). If absent, create it when you materialize the manifest.

Read the full file on GitHub · 391 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 391 lines · 79 tokens per session scan A fe31550572b1

Subscribe to this mod's changes

parallel-dispatcher is an agent published in the GitHub repository procoders/superpowers-v (35 stars, last pushed 3d ago), licensed MIT. It adds 79 tokens to every session and 12,034 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.