run-plan

A command that reads an approved implementation plan and starts a workflow that executes independent repository tasks in parallel. It requires a plan path, repository path, and integration branch when these are not unambiguous.

In plain words
What is it for?
Use it to execute a parsed implementation plan across a target repository, resume remaining tasks after an interruption, or start the plan on a chosen integration branch.
Why use it?
It reduces manual coordination when a plan contains tasks that can run independently. It also checks for unfinished state from an earlier interrupted run before continuing.

Command

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 commands/bacsystem/parallel-plan-executor/run-plan
Clone the repo
git clone --depth 1 https://github.com/bacsystem/parallel-plan-executor
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,349 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 $0.00017 $0.02349
Opus 5 $0.00009 $0.01175
Sonnet 5 $0.00003 $0.00470
Haiku 4.5 $0.00002 $0.00235

Measured yesterday against content hash 2bfbe2019fbe, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

run-plan 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 yesterday.

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.

commands/run-plan.md · 161 lines

How it starts

The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.

What this command does

Runs the parallel-plan-executor Workflow against a cys:plan implementation plan, executing its independent tasks in parallel instead of one at a time.

REPO = ${CLAUDE_PLUGIN_ROOT}

Steps

  1. Parse $ARGUMENTS as up to three whitespace-separated tokens: plan-path, repo-path, integration-branch. Any of the three that's missing or ambiguous — ask the user for it in plain language before continuing. Do not guess a plan path, a target repo, or a branch name.

  2. Check for leftover state from an interrupted run: check whether <repo-path>/.cys/state.json exists. If it does not, continue normally — step 4 below uses bin/parse-plan.js as usual.

    • If it exists, read it:
      • If its planPath matches <plan-path> exactly: tell the user which tasks are already done/failed/pending per the file, and ask whether to continue with only what's left (step 4 below then uses bin/plan-remainder.js instead of bin/parse-plan.js) or start fresh (delete <repo-path>/.cys/state.json first; step 4 then uses bin/parse-plan.js as usual).
      • If its planPath does not match: warn the user there's incomplete state from a different, unrelated run and ask how to proceed (look at it first / delete it and continue / stop) — never decide silently.
  3. Sanity-check before running anything:

    • Confirm plan-path exists and looks like an approved plan (has ### Task N: blocks). If it looks like a spec instead of a plan, say so and stop.
    • Confirm repo-path is a git repo with a clean working tree (git status). If it's dirty, tell the user and stop — don't run against uncommitted work.
    • Confirm integration-branch is an ephemeral feature branch (e.g. feature/<name>), not develop/main/master directly. If the user asked for one of those directly, warn them (mainline should never take agent merges directly) and confirm they really want that before proceeding.

Read the full file on GitHub · 161 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. yesterday First seen · 161 lines · 0 tokens per session scan A 2bfbe2019fbe

Subscribe to this mod's changes

run-plan is a command published in the GitHub repository bacsystem/parallel-plan-executor (2 stars, last pushed 1mo ago), licensed MIT. It adds 17 tokens to every session and 2,349 once invoked, about $0.0001 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 2026-08-31.