docs: add task-reviewer agent to CUSTOM-AGENTS.md (#363)

littleKitchen · WilliamBerryiii · web-flow · commit 0efb72211a3d · 2026-01-31T12:53:12.000-08:00
## Summary Fixes #310 The `task-reviewer.agent.md` exists in `.github/agents/` and has comprehensive RPI documentation at `docs/rpi/task-reviewer.md`, but was missing from the main agent reference document. ## Changes - Added `task-reviewer` to the RPI Workflow Agents table after `task-implementor` - Added Agent Details section following existing pattern: - **Creates:** Review validation logs - **Workflow:** Locate artifacts → Extract checklist → Validate items → Run commands → Document findings - **Critical:** Review-only specialist with severity levels ## References - Agent file: `.github/agents/task-reviewer.agent.md` - RPI documentation: `docs/rpi/task-reviewer.md` Co-authored-by: Bill Berry <WilliamBerryiii@users.noreply.github.com>
diff --git a/.github/CUSTOM-AGENTS.md b/.github/CUSTOM-AGENTS.md
@@ -44,6 +44,7 @@ The Research-Plan-Implement (RPI) workflow provides a structured approach to com
 | **task-researcher** | Produces research documents with evidence-based recommendations | Research-only; never plans or implements |
 | **task-planner** | Creates 3-file plan sets (plan, details, prompt) | Requires research first; never implements code |
 | **task-implementor** | Executes implementation plans with subagent delegation | Requires completed plan files |
+| **task-reviewer** | Validates implementation against research and plan specifications | Requires research/plan artifacts |
 
 ### Documentation and Planning Agents
 
@@ -128,6 +129,18 @@ The Research-Plan-Implement (RPI) workflow provides a structured approach to com
 
 **Critical:** Requires completed plan files. Uses subagent architecture for parallel phase execution. Updates tracking artifacts after each phase.
 
+### task-reviewer
+
+**Creates:** Review validation logs:
+
+* `.copilot-tracking/reviews/{{YYYY-MM-DD}}-<topic>-review.md`
+
+**Workflow:** Locate artifacts → Extract checklist → Validate items → Run commands → Document findings
+
+**Critical:** Review-only specialist. Validates against documentation, not assumptions. Produces findings with severity levels (Critical, Major, Minor).
+
+**Documentation:** See [Task Reviewer Guide](../docs/rpi/task-reviewer.md) for detailed usage.
+
 ### prompt-builder
 
 **Creates:** Instruction files and prompt files:
