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.
npx agentmods add commands/transparent-pegasus/herdrpowers/plangit clone --depth 1 https://github.com/transparent-pegasus/herdrpowersWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/transparent-pegasus/herdrpowers/plan)<a href="https://agentmods.dev/commands/transparent-pegasus/herdrpowers/plan"><img src="https://agentmods.dev/badge/commands/transparent-pegasus/herdrpowers/plan.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00011 | $0.01064 |
| Opus 5 | $0.00005 | $0.00532 |
| Sonnet 5 | $0.00002 | $0.00213 |
| Haiku 4.5 | $0.00001 | $0.00106 |
Grade A, and why
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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/plan Workflow
This command runs only the discovery and planning stages of the development cycle. The plan is not complete when it is written — it is complete after independent review, resolution of the findings, and the user's approval.
Deliverable Saves
Whenever this workflow saves a deliverable, also save any corresponding design/plan files in the same change so the artifact and its rationale stay together.
Workflow File Edits
Do not edit workflow files unless the user's current request explicitly asks to change a workflow file or workflow behavior. A workflow path alone can select the workflow context; it is not permission to edit that file. For example, update commands/plan.md ... permits editing the workflow, while commands/plan.md add a feature ... to the app does not.
Configuration
<KEY> placeholders in this workflow (for example <PLAN_PATH_PATTERN>) resolve from the Herdrpowers Configuration section of the repository's instruction files (CLAUDE.md / AGENTS.md). If that section is missing, run the pack's init workflow first: /herdrpowers:init on Claude Code plugin installs, or commands/init.md otherwise.
Orchestration
Planning and design stay with the orchestrator — the pane the user typed the request into. Only the reviews are delegated, through the orchestration skill with using-herdr-sibling-panes as the transport. Read both before the first delegation, and resolve the repository's assignments and review gates as orchestration describes: .herdrpowers/config.yaml first, then orchestration/roles.yaml for anything it omits. Every gate this workflow would run is resolved up front; a gate set to enabled: false is skipped and named in the final report. In-process subagents are not used for workflow work, with one exception: a review task whose resolved mode is orchestrator runs in a fresh subagent the orchestrator spawns.
Execution Steps
- Brainstorming Ask the user what they want to build. Read the brainstorming SKILL.md. Invoking this workflow is the user choosing its ceremony: announce the brainstorming skill's classification, but do not downgrade the workflow — its remaining steps run even for bounded work. If the request is a spike (the output is an answer, not code to keep), say so and confirm with the user before continuing. Engage in a design and requirement gathering discussion without writing implementation code. Proceed to Step 2 as soon as the last clarifying question has been answered and there are no unresolved design concerns.
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.
- 4d ago First seen · 61 lines · 11 tokens per session scan A d36cef5da513
plan is a command published in the GitHub repository transparent-pegasus/herdrpowers (11 stars, last pushed 7d ago), licensed Apache-2.0. It adds 11 tokens to every session and 1,064 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 2026-08-30.
Other commands, from other repositories
v-epic
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…
v-init
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.
v-models
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…
v-collect
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.
v-status
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.
v-orchestrate
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…