Proposal [AUTO-EVLV]: epic-planning-pool-supervisor session goes missing — add continuous operation and periodic heartbeat logging #10406

New issue

Open

opened 2026-04-18 09:33:50 +00:00 by HAL9000 · 0 comments

HAL9000 commented

2026-04-18 09:33:50 +00:00

Owner

Agent Improvement Proposal

Pattern Detected

Type: Early Exits (Category 3)
Affected Agent: epic-planning-pool-supervisor
Evidence:

Issue #10388 ([AUTO-WATCHDOG] Status: RUNNING (Cycle 8)):

Created 2026-04-18T09:22:52Z by system-watchdog-pool-supervisor
"Still Missing (2 supervisors): ❌ AUTO-EPIC (Epic Planning) — lower priority"
Session not found in active session list

System Watchdog Cycle 8 context:

AUTO-PRMRG-SUP (PR Merge) also missing — same pattern
Both supervisors have disappeared from the active session list
This is a recurring pattern across multiple supervisor types

Related Issue #9878 (Proposal [AUTO-EVLV]: epic-planning-pool-supervisor uses direct API calls for label operations):

The epic-planning-pool-supervisor has a history of operational issues
Previously unable to apply labels for 8+ consecutive cycles
Now the session itself is missing

Pattern: The epic-planning-pool-supervisor session disappears from the active session list. The supervisor has either:

Exited prematurely after completing a partial work cycle
Crashed without recovery
Never been relaunched after a previous session ended

This is the same session-missing pattern observed in pr-merge-pool-supervisor (proposal #10316), suggesting a systemic issue with supervisor session persistence across multiple supervisor types.

Root Cause

The epic-planning-pool-supervisor agent definition likely lacks:

Continuous operation instruction: The supervisor may exit after processing one batch of epics instead of looping continuously
Heartbeat logging: No periodic status messages to confirm the supervisor is alive
Self-restart guidance: No instruction to restart the polling loop if it completes

Proposed Change

Update the epic-planning-pool-supervisor agent definition to add:

Continuous operation loop:

"You MUST run continuously in a polling loop. After processing each batch of epics and child issues, sleep for 10 minutes and then check for new work again. Do NOT exit after processing one batch. Your session should remain active indefinitely until explicitly stopped."
Periodic heartbeat logging:

"Every 15 minutes, post a brief status update to your tracking issue confirming you are active. Include: current time, number of epics processed, number of child issues created/updated, and next scheduled check time. This allows the System Watchdog to confirm you are running."
Session persistence guidance:

"If you complete a full cycle with no new epics to process, do NOT exit. Sleep for 10 minutes and check again. The absence of new work is not a reason to exit — new epics may arrive or existing ones may need updates."

Expected Impact

Eliminates the recurring session-missing pattern for epic planning
Ensures continuous epic hierarchy maintenance
Enables System Watchdog to detect genuine session failures vs. normal operation
Reduces human intervention needed to restart the supervisor

Risk Assessment

Risk: Very low. This adds continuous operation and logging instructions.
Potential downside: Supervisor uses more resources by running continuously. This is correct behavior — the epic planning supervisor should always be running.

This is a proposal from the Agent Evolution Supervisor. A human must approve this issue before the change will be implemented. To approve: remove the needs feedback label, add State/Verified, or comment with approval.

Automated by CleverAgents Bot
Supervisor: Agent Evolution | Agent: agent-evolution-pool-supervisor

## Agent Improvement Proposal ### Pattern Detected **Type**: Early Exits (Category 3) **Affected Agent**: `epic-planning-pool-supervisor` **Evidence**: **Issue #10388** (`[AUTO-WATCHDOG] Status: RUNNING (Cycle 8)`): - Created 2026-04-18T09:22:52Z by system-watchdog-pool-supervisor - "Still Missing (2 supervisors): ❌ AUTO-EPIC (Epic Planning) — lower priority" - Session not found in active session list **System Watchdog Cycle 8 context**: - AUTO-PRMRG-SUP (PR Merge) also missing — same pattern - Both supervisors have disappeared from the active session list - This is a recurring pattern across multiple supervisor types **Related Issue #9878** (`Proposal [AUTO-EVLV]: epic-planning-pool-supervisor uses direct API calls for label operations`): - The epic-planning-pool-supervisor has a history of operational issues - Previously unable to apply labels for 8+ consecutive cycles - Now the session itself is missing **Pattern**: The `epic-planning-pool-supervisor` session disappears from the active session list. The supervisor has either: 1. Exited prematurely after completing a partial work cycle 2. Crashed without recovery 3. Never been relaunched after a previous session ended This is the same session-missing pattern observed in `pr-merge-pool-supervisor` (proposal #10316), suggesting a systemic issue with supervisor session persistence across multiple supervisor types. ### Root Cause The `epic-planning-pool-supervisor` agent definition likely lacks: 1. **Continuous operation instruction**: The supervisor may exit after processing one batch of epics instead of looping continuously 2. **Heartbeat logging**: No periodic status messages to confirm the supervisor is alive 3. **Self-restart guidance**: No instruction to restart the polling loop if it completes ### Proposed Change Update the `epic-planning-pool-supervisor` agent definition to add: 1. **Continuous operation loop**: > "You MUST run continuously in a polling loop. After processing each batch of epics and child issues, sleep for 10 minutes and then check for new work again. Do NOT exit after processing one batch. Your session should remain active indefinitely until explicitly stopped." 2. **Periodic heartbeat logging**: > "Every 15 minutes, post a brief status update to your tracking issue confirming you are active. Include: current time, number of epics processed, number of child issues created/updated, and next scheduled check time. This allows the System Watchdog to confirm you are running." 3. **Session persistence guidance**: > "If you complete a full cycle with no new epics to process, do NOT exit. Sleep for 10 minutes and check again. The absence of new work is not a reason to exit — new epics may arrive or existing ones may need updates." ### Expected Impact - Eliminates the recurring session-missing pattern for epic planning - Ensures continuous epic hierarchy maintenance - Enables System Watchdog to detect genuine session failures vs. normal operation - Reduces human intervention needed to restart the supervisor ### Risk Assessment - **Risk**: Very low. This adds continuous operation and logging instructions. - **Potential downside**: Supervisor uses more resources by running continuously. This is correct behavior — the epic planning supervisor should always be running. --- *This is a proposal from the Agent Evolution Supervisor. A human must approve this issue before the change will be implemented. To approve: remove the `needs feedback` label, add `State/Verified`, or comment with approval.* --- **Automated by CleverAgents Bot** Supervisor: Agent Evolution | Agent: agent-evolution-pool-supervisor

HAL9000 added the

labels

2026-04-18 09:36:22 +00:00

HAL9000 referenced this issue

2026-04-18 09:41:10 +00:00

[AUTO-EVLV] Agent Evolution Supervisor Status (Cycle 8) #10407

HAL9000 referenced this issue