Plugin

guild

Guild turns Claude Code into a self-evolving team of specialist agents for larger software tasks. It plans before execution, selects the right specialists, assembles focused context, preserves durable project memory, and improves its skills over time through evidence-based reflection. Instead of treating every request as a single long prompt, Guild breaks work into a governed workflow: clarify the goal, compose the right agent team, execute with scoped context, review the results, capture decisions, and only evolve specialists or skills when repeated evidence shows they are useful. When native agent teams are available, Guild coordinates them directly; when they are not, it falls back to sub-agents with the same planning and memory discipline. Guild is designed for projects where context, quality, and continuity matter: feature work, refactors, documentation, QA, release preparation, research-heavy implementation, and recurring workflows that benefit from reusable specialist knowledge.

• Published source
• Repository

Analysed

11 October 2026

Last observed

11 October 2026

What the code does

• Runs shell commands Code that starts system commands on the machine it runs on.
• Starts other processes Code that launches separate programs.
• Downloads and runs code Code that fetches a script and pipes it into a shell.
• Evaluates code at runtime Code that builds and runs code while it executes.
• Runs hooks automatically Hooks that execute without being called explicitly.
• Runs an AI coding agent unsupervised Code that starts an AI coding agent with its approval prompts switched off.
• Reads AI and developer logins Code that refers to AI tool credentials or publishing tokens, such as .npmrc.
• Asks your agent's AI to write text The server can request text from the agent's own model (MCP sampling).

Having a capability is not wrongdoing; many tools need these to do their job. It tells you what to check before you install.

This is public metadata, not a safety certification or complete assessment. Static analysis cannot establish what a tool does at runtime.