init

A setup command that records repository-specific settings for the Herdrpowers workflow in the repository's instruction files and configuration. It assigns roles to tasks such as implementation, testing, verification, and review.

In plain words
What is it for?
Initialize a repository, create its .herdrpowers/config.yaml file, declare required settings, and prepare those changes to be committed.
Why use it?
It lets shared workflow instructions use the correct paths, commands, and task assignments for a particular repository without changing the shipped workflow files.

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/transparent-pegasus/herdrpowers/init
Clone the repo
git clone --depth 1 https://github.com/transparent-pegasus/herdrpowers
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 4,196 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 $0.04196
Opus 5 $0.00009 $0.02098
Sonnet 5 $0.00003 $0.00839
Haiku 4.5 $0.00002 $0.00420

Measured 2d ago against content hash 0705dbb83ea5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

init 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.

commands/init.md · 246 lines

How it starts

The opening of the file, as written. The whole thing — 246 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/init Workflow (herdrpowers)

herdrpowers skills and commands reference repo-specific values through <KEY> placeholders (for example <PLAN_PATH_PATTERN>, <BASELINE_VERIFICATION_COMMAND>). This workflow declares the actual values inside the target repository's own instruction files, where every pack file resolves them at runtime. Pack files themselves are never edited: plugin installations are read-only, and checked-in copies stay diff-clean against the pack.

It also writes .herdrpowers/config.yaml: the repository's role list and the role assigned to every delegation task — implementation, tests, chores, verification, and each review. That file is the only place a repo customizes who takes which task and where it runs; skills/orchestration/roles.yaml stays untouched as the shipped default.

Deliverable Saves

This workflow has two deliverables:

  1. The Herdrpowers Configuration section written into the repository's instruction files.
  2. .herdrpowers/config.yaml at the repository root — the role list and the per-task role assignments.

Commit both with the repository's normal instruction-file conventions. .herdrpowers/config.yaml is tracked configuration, not scratch — do not add it to .gitignore. (The sibling .herdrpowers/pdd/ workspace ignores itself and is unaffected.)

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.

Execution Steps

  1. Inventory Inspect the repository to ground a proposal for every key below: build and test tooling (package.json, Makefile, pyproject.toml, Cargo.toml, CI workflow files), docs layout, and existing CLAUDE.md / AGENTS.md content. If a Herdrpowers Configuration section already exists in an instruction file, load it and treat this run as an update of that section. Read the pack's skills/orchestration/roles.yaml for the shipped role list, the shipped delegation: defaults, and the default role assigned to each delegation task. If .herdrpowers/config.yaml already exists, load it too and treat this run as an update of both deliverables — never silently reset a customized assignment or a gate the user turned off. If herdr is running (HERDR_ENV=1), run herdr pane list and note which agent types actually exist in the session, and which of them share the caller's tab_id. Delegation is confined to the orchestrator's own tab by default, so an agent type that exists only in another tab is not usable by this repository's workflows as configured — flag that, alongside any default assignment naming an agent type this user does not have at all.

Read the full file on GitHub · 246 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 · 246 lines · 17 tokens per session scan A 0705dbb83ea5

Subscribe to this mod's changes

init is a command published in the GitHub repository transparent-pegasus/herdrpowers (11 stars, last pushed 5d ago), licensed Apache-2.0. It adds 17 tokens to every session and 4,196 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.