Enforceable, provider-neutral harness for multi-agent AI coding: spec-driven workflow, cross-provider peer review, and safety rails. First-class for Claude Code and Pi; advisory adapters for Cursor, Codex, Copilot, Windsurf, and Aider.
This repository also configures its own agents. See what code-copilot-team tells them →
Hook Claude Code
One of 8 in settings.json
Runs before the agent uses a tool for Edit, Write and Bash tool calls, executing protect-files.sh and protect-git.sh (2 commands). From gosha70/code-copilot-team.
Hook Claude Code
One of 8 in settings.json
Runs when the agent finishes a response, executing verify-on-stop.sh and peer-review-on-stop.sh (2 commands). From gosha70/code-copilot-team.
Hook Claude Code
One of 8 in settings.json
Runs after a tool call finishes for Edit and Write tool calls, executing verify-after-edit.sh and auto-format.sh (2 commands). From gosha70/code-copilot-team.
Hook Claude Code
One of 8 in settings.json
Runs when the agent sends a notification, executing notify.sh. From gosha70/code-copilot-team.
Hook Claude Code
One of 8 in settings.json
Runs when a session starts, executing reinject-context.sh, memkernel-recall.sh and harness-stamp.sh (3 commands). From gosha70/code-copilot-team.
Hook Claude Code
One of 8 in settings.json
Runs before the context is compacted, executing memkernel-pre-compact.sh. From gosha70/code-copilot-team.
Hook Claude Code
One of 8 in settings.json
Runs after the context is compacted, executing memkernel-post-compact.sh. From gosha70/code-copilot-team.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: