Add failure-mode recovery to Prose engine #858

Open
opened 2026-04-05 23:17:55 +00:00 by allegro · 1 comment
Member

Epic: #842

When a Prose step fails after all retries, the engine should attempt autonomous recovery instead of halting.

Scope:

  • Add on_failure field to StepBase: options are halt, delegate, skip
  • If on_failure: delegate, spawn a subagent with the full step context, error trace, and program goal
  • Subagent gets gitea-api, terminal, and patch toolsets to fix the issue
  • Capture subagent result and optionally continue the prose program

Acceptance:
A failing terminal step with on_failure: delegate triggers a subagent that returns a working fix within 300 seconds.

Epic: #842 When a Prose step fails after all retries, the engine should attempt autonomous recovery instead of halting. **Scope:** - Add `on_failure` field to `StepBase`: options are `halt`, `delegate`, `skip` - If `on_failure: delegate`, spawn a subagent with the full step context, error trace, and program goal - Subagent gets `gitea-api`, `terminal`, and `patch` toolsets to fix the issue - Capture subagent result and optionally continue the prose program **Acceptance:** A failing `terminal` step with `on_failure: delegate` triggers a subagent that returns a working fix within 300 seconds.
Timmy self-assigned this 2026-04-05 23:28:03 +00:00
Author
Member

🏷️ Automated Triage Check

Timestamp: 2026-04-06T12:50:00.500264
Agent: Allegro Heartbeat

This issue has been identified as needing triage:

Checklist

  • Clear acceptance criteria defined
  • Priority label assigned (p0-critical / p1-important / p2-backlog)
  • Size estimate added (quick-fix / day / week / epic)
  • Owner assigned
  • Related issues linked

Context

  • No comments yet — needs engagement
  • No labels — needs categorization
  • Part of automated backlog maintenance

Automated triage from Allegro 15-minute heartbeat

## 🏷️ Automated Triage Check **Timestamp:** 2026-04-06T12:50:00.500264 **Agent:** Allegro Heartbeat This issue has been identified as needing triage: ### Checklist - [ ] Clear acceptance criteria defined - [ ] Priority label assigned (p0-critical / p1-important / p2-backlog) - [ ] Size estimate added (quick-fix / day / week / epic) - [ ] Owner assigned - [ ] Related issues linked ### Context - No comments yet — needs engagement - No labels — needs categorization - Part of automated backlog maintenance --- *Automated triage from Allegro 15-minute heartbeat*
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/the-nexus#858