deadfish-cli

Orchestrate autonomous development pipelines with CLAUDE Code and deterministic verification.

Updated Jan 28, 2026
One-click install
npx skills add https://github.com/Fredasterehub/deadfish-cli --skill deadfish-cli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deadfish-cli
Source: https://github.com/Fredasterehub/deadfish-cli/tree/main
Command: npx skills add https://github.com/Fredasterehub/deadfish-cli --skill deadfish-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bash, python3, yq, codex, claude.

What problem does it solve?

The deadfish-cli project implements an autonomous development pipeline that plans, codes, verifies, and commits changes with no human intervention, enabling teams to push updates while focusing on higher-level work.

Core Features & Use Cases

  • End-to-end automation: from task planning to code commits with deterministic verification.
  • Living docs: generates and maintains VISION.md, PROJECT.md, REQUIREMENTS.md, ROADMAP.md, and STATE.yaml artifacts.
  • Safe, auditable workflow: strict role boundaries, atomic state updates, and transparent logging.

Quick Start

Run the pipeline in a prepared project: configure STATE.yaml and POLICY.yaml, then execute the kickoff script.

Frequently Asked Questions about deadfish-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate a development pipeline from task planning to code commits?

You can automate a development pipeline by configuring STATE.yaml and POLICY.yaml in a prepared project, then executing the kickoff script to orchestrate planning, coding, verification, and commits autonomously.

What are living docs in an autonomous development workflow?

Living docs are continuously maintained artifacts—VISION.md, PROJECT.md, REQUIREMENTS.md, ROADMAP.md, and STATE.yaml—that an autonomous pipeline generates and updates to track project state and guide future work.

Do I need bash and python3 to run an autonomous coding pipeline?

Yes, running this autonomous coding pipeline requires bash, python3, yq, codex, and claude to handle CLI orchestration, planning, implementation, configuration parsing, and verification gates.

How does a deterministic verifier work in an automated code generation pipeline?

A deterministic verifier enforces strict quality gates in an automated pipeline by validating generated code against predefined rules before atomic commits, ensuring safe and auditable workflow boundaries.

Can I use CLAUDE Code for orchestration in an unattended CI/CD pipeline?

Yes, CLAUDE Code can orchestrate an unattended CI/CD pipeline by coordinating planning via Codex and enforcing deterministic verification gates, enabling autonomous commits while maintaining transparent logging and strict role boundaries.