agentmods.dev

init

A setup command for choosing Claude models and reasoning levels for the planning, checking, and coding roles. Claude is an AI assistant, and a subscription or API key controls which models you can use.

In plain words
What is it for?
Use it to create or update the model-routing configuration for agent-harness and review an existing setup.
Why use it?
It removes the assumption that every user has access to the same Claude model and updates older configuration formats when needed.

Command

Part of the agent-harness plugin — 5 skills, 1 command, 1 hook shipped together

5 1mo ago A scan 15 tokens original MIT
View the source file bouob/agent-harness
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/bouob/agent-harness/init
Clone the repo
git clone --depth 1 https://github.com/bouob/agent-harness

Or install agent-harness, the plugin that ships this one along with the rest of its 5 skills, 1 command, 1 hook.

Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,167 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 $0.00015 $0.03167
Opus 5 $0.00008 $0.01584
Sonnet 5 $0.00003 $0.00633
Haiku 4.5 $0.00002 $0.00317

Measured 3d ago against content hash 726f2b17e65d, 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 3d 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 · 313 lines

How it starts

The opening of the file, as written. The whole thing — 313 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/agent-harness:init — Configure Model Routing

Walk the user through writing ~/.claude/agent-harness.json so /sprint knows which Claude models and reasoning effort to assign to each role (Planner, Evaluator, Generator). Defaults assume Opus access; this wizard lets users on Pro/Team subscriptions, Sonnet-only API keys, or any other access shape route around the assumption.

v2.3.0 added per-role effort (reasoning level). Each role is now {model, effort} instead of a bare model string. v1 / v2 / v3 configs auto-lift to v4 on first read.

Schema reference: ${CLAUDE_PLUGIN_ROOT}/skills/sprint/references/config-schema.md.

Step 1 — Detect Existing Config

Try Read on ~/.claude/agent-harness.json. If the file exists:

  1. Parse the JSON. If version is missing or 1 / 2 / 3, run the auto-lift described in references/config-schema.md § Migration. v2 configs whose any models.<role>.engine != "claude" ABORT here with:

    "v2 config has engine= for role . v0.6.0 dropped multi-host support. Reconfigure?" and offer Reconfigure / Cancel.

  2. Show the user the current models block in a fenced code block.
  3. Use AskUserQuestion: "Existing config found. Reconfigure or keep?"
    • Reconfigure → continue to Step 2
    • Show current and exit → print the parsed config, stop
    • Cancel → stop without printing

If the file does not exist, proceed straight to Step 2.

Step 2 — Ask Which Models the User Has Access To

Use AskUserQuestion to ask: "Which Claude models can you use? (works for both Claude.ai subscriptions and direct API access)"

Use the preview field on each option to show a markdown table of the resulting routing — this lets the user compare presets visually without reading JSON.

Options (display text → internal preset → preview content):

Option 1 — Fable 5 + all models → frontier

preview:

| Role                | Model  | Effort |
|---------------------|--------|--------|
| Planner             | Fable  | high   |
| Evaluator           | Opus   | high   |
| Generator (code)    | Sonnet | high   |
| Generator (write)   | Sonnet | high   |
| Generator (research)| Sonnet | high   |
| Generator (collect) | Haiku  | —      |

Note: Fable 5 planner costs ~2x the Opus planner. Haiku takes no effort.

Read the full file on GitHub · 313 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. 3d ago First seen · 313 lines · 15 tokens per session scan A 726f2b17e65d

Subscribe to this mod's changes

init is a command published in the GitHub repository bouob/agent-harness (5 stars, last pushed 1mo ago), licensed MIT. It adds 15 tokens to every session and 3,167 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.

AutonomousMulti-Agentagent-teamsclaude-codeclaude-code-pluginharness-engineering
Related

Other commands, from other repositories

init

First-run teamctl onboarding — from no-teamctl-installed to a running supervised team in one conversation.

Alireza29675/teamctl · 20 tokens

create

Scaffold a new plugin with directory structure, manifests, and marketplace registration.

rube-de/cc-skills · 14 tokens

adjust

Open-ended "talk to it" command for evolving an existing teamctl team — add or remove an agent session, modify its behaviour or capability stack (sub-agents / skills / /loop / hook), scope a channel, set up Telegram, add a project, or open a bridge.

Alireza29675/teamctl · 59 tokens

bugfix

TDD-driven bugfix workflow: tester writes failing test (RED) → developer fixes (GREEN) → developer refactors (REFACTOR) → reviewer validates. Accepts issue number, description, or both. Auto-creates PR unless --no-pr flag is passed.

rube-de/cc-skills · 55 tokens

develop

Implement skill development issues with TDD-governed workflow.

rube-de/cc-skills · 11 tokens

dev-task

Create an agent team to develop: Developer teammate + Code-tester teammate + QA-tester teammate + Reviewer teammate + Researcher subagent → implements .dev/cdt/plans/plan-$TIMESTAMP.md in waves.

rube-de/cc-skills · 43 tokens