28 tagged agent pipeline, measured the same way as everything else here.
Browse within: Multi-Agent 15llm-tools 15bun 12game 12
Agent Claude Code
Final review of the diff for correctness, security, conventions, and spec conformance. Waits for CI green. On clean verdict, appends one line to LEARNINGS.md and signals merge-ready. On blocking findings, emits HANDOFF:FIX to coder. Read-only on source code.
Agent
Implement a plan from IssuePlanner, run Bastion formatters and linters, commit, and open a PR.
Agent
Read-only reviewer. Reads the PR diff, identifies issues, and hands off to IssueCoder to fix them. Never writes code.
Agent
Golden CodeBuddy reviewer.
Agent
Diagnostic specialist — tries to break the proposed fix before it ships. Red-team the assumptions. Read-only investigator. Does NOT write fixes. Spawned by /brigade:bug-council or /brigade:fix on 3+ failed fix attempts.
Agent
Sprint planner — breaks a sprint goal into atomic task specs with file ownership, dependencies, and waves for parallel execution. Does NOT execute tasks and does NOT spawn worker agents. Use when: planning a new sprint, decomposing a goal into tasks, assigning waves. Triggers on: "plan sprint", "create task specs"…
Agent
Diagnostic specialist — finds similar bugs and similar patterns elsewhere in the codebase. If the same bug exists in 5 places, the fix should probably be systemic. Read-only. Spawned by /brigade:bug-council or /brigade:fix on 3+ failed fix attempts.