hermes/agent-framework
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add Test Design phase: test_design.md prompt, workflow state machine, orchestrator updates, implement/referee reference TEST_PLAN.md

Browse Source
This commit is contained in:
gitea
2026-06-11 09:25:45 -04:00
parent fec11d29dc
commit f4587886b9
10 changed files with 193 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
prompts/workflow.md
+3 -2
View File
@@ -8,7 +8,8 @@ This file defines the linear progression of a task in the agent-framework. The O
| :--- | :--- | :--- | :--- |
| **New Task** | Orchestrator creates `tasks/{task-name}/` with **no files** | Research | Generate `SPEC.md` |
| **Research** | Has `SPEC.md` | Design (optional) or Implement | Generate `DESIGN.md` or code |
| **Design** | Has `DESIGN.md` | Implement | Generate code and tests |
| **Design** | Has `DESIGN.md` | Test Design (optional) or Implement | Generate `TEST_PLAN.md` or code |
| **Test Design** | Has `TEST_PLAN.md` | Implement | Generate code and tests |
| **Implementation** | Has `IMPLEMENTATION.md` | Bug Find | Generate `BUG_REPORT.md` |
| **Bug Find** | Has `BUG_REPORT.md` | Adversarial Bug Find | Generate `ADVERSARIAL_BUG_REPORT.md` |
| **Adversarial Bug Find** | Has `ADVERSARIAL_BUG_REPORT.md` | Doc Review | Generate `DOC_REVIEW.md` |
@@ -58,7 +59,7 @@ When the Orchestrator detects a `VERDICT.md` with `FAIL` or `NEEDS_REVIEW`, the
## Autopilot Rules
1. **Linear Progression**: Never skip a phase (except Design, which is optional).
1. **Linear Progression**: Never skip a phase (except Design and Test Design, which are optional).
2. **Artifact Check**: A phase is only considered "complete" if its corresponding artifact exists and is non-empty.
3. **Automatic Transition**: Upon completion of an artifact, the Orchestrator must immediately identify and propose the next phase in the lifecycle.
4. **Human Intervention**: If the Referee marks a task as `FAIL`, `NEEDS_REVIEW`, or identifies "Tie-Breaks", the Autopilot pauses and waits for user input — the Orchestrator does NOT auto-create fix/review/tiebreak tasks in Autopilot mode; the user must decide whether to create fix tasks and how to proceed. (In manual mode, the Orchestrator auto-creates these tasks.)