Agents
Reference documentation for the agents shipped by fullsend. The default agents are defined in the fullsend-ai/agents repository and resolved at runtime via config entries or agents-repo fallback. Custom agents can be registered via the agents: field in org or per-repo config (see Architecture for details on the registration model).
| Agent | Summary |
|---|---|
| Triage | Inspects new issues and produces structured triage decisions |
| Prioritize | Scores issues using the RICE framework for project board ranking |
| Code | Implements fixes and features from triaged issues |
| Review | Reviews pull requests for correctness, security, and intent alignment |
| Fix | Addresses review feedback on open PRs |
| Retro | Analyzes completed workflows and proposes system improvements |
Configuration
All agents can be configured by adding instructions and skills to your repository. Changes to AGENTS.md affect every agent; skills let you tune how a specific agent performs a specific task. See Configuring with AGENTS.md and Configuring with Skills.
At some point, enough configuration turns a configured default agent into a derived agent. See Default, derived, and custom agents for where that line is and why it matters.
Custom Agents
Custom agents can be added to the fullsend pipeline via the agents: field in your org-level or per-repo config.yaml. Each entry is either a local path (relative to the fullsend directory) or a pinned HTTPS URL with an integrity hash. Config entries are looked up directly by name; when absent, a runtime fallback resolves known first-party agents from fullsend-ai/agents. See Bring Your Own Agent for the complete guide to building and registering custom agents.