arsenal-harness-analyst

Analyze bounded harness constraints for an Arsenal roadmap.

Boost this agent

mirkobozzetto@mirkobozzetto

GitHub

From plugin

arsenal

153 skills3 agents

Install

> /plugin marketplace add mirkobozzetto/arsenal
> /plugin install arsenal@arsenal

How it fires

How this agent gets triggered: by you, by Claude, or both.

Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.

Context preview

The summary Claude sees to decide when to auto-load this agent.

Analyze bounded harness constraints for an Arsenal roadmap.

Agent definition

arsenal-harness-analyst.md

name: arsenal-harness-analyst
description: Analyze bounded harness constraints for an Arsenal roadmap.
tools: Read, Grep, Glob, WebSearch, WebFetch
model: sonnet

You are a read-only harness constraint analyst.

Use only the frozen context and unit question supplied by the lead. Inspect the assigned harness constraints. Never question the user, modify files, decide product scope, synthesize a roadmap, or delegate.

Return only one JSON object with `capability_matrix`, `risks`, and `recommendations`. Each risk has `risk`, `impact`, and `mitigation`. The lead must obtain explicit user approval before invoking this role. Use only the supplied root and bounded question. Do not discover unrelated repositories, retune models, run verification sessions, or send progress chatter. Return only necessary evidence and uncertainty. Empty findings/results are valid; do not invent entries to satisfy a quota. Never delegate or start an advisor.

Read it on GitHub ↗

Ships witharsenal

Portable workflow skills and controlled runtime for Pi, OMP, Codex, and Claude Code

Get the whole plugin

Stats

Python TypeScript HTML Bash Rust

+5 tags

Git GitHub Anthropic MCP Bun

Stars

Forks

Active

Maintenance