Rockachopa/hermes-config
1
0
Fork 0
Code Issues 4 Pull Requests Actions Packages Projects Releases Wiki Activity

[OPS] Implement use-it-or-lose-it quota utilization tracking for fleet workers #4

New Issue
Open
opened 2026-04-05 19:41:58 +00:00 by Timmy · 1 comment
Timmy commented 2026-04-05 19:41:58 +00:00
Copy Link

Problem

Current agent fleet utilization is invisible. We need a utilization scoreboard to track tokens consumed vs output produced per agent per day.

What To Do

Create a lightweight utilization tracker script:

  1. Read loop log files in ~/.hermes/logs/
  2. Count: commits pushed, PRs filed, PRs merged per agent in last 24h
  3. Check: is the agent loop process alive?
  4. Check: any Gitea activity from that user in last 24h
  5. Write summary to ~/.hermes/logs/utilization-report.json
  6. If 0 output for 12+ hours, flag as ZOMBIE; if quota available but low utilization, flag as IDLE

Acceptance Criteria

  • Script exists at ~/.hermes/bin/utilization-check.sh
  • Script outputs per-agent status table when run manually
  • Script can be called from cron every hour
  • ZOMBIE/IDLE flags are accurate
  • Comment on this issue with first-run output

Assignee: groq
Labels: ops, priority-medium

## Problem Current agent fleet utilization is invisible. We need a utilization scoreboard to track tokens consumed vs output produced per agent per day. ## What To Do Create a lightweight utilization tracker script: 1. Read loop log files in ~/.hermes/logs/ 2. Count: commits pushed, PRs filed, PRs merged per agent in last 24h 3. Check: is the agent loop process alive? 4. Check: any Gitea activity from that user in last 24h 5. Write summary to ~/.hermes/logs/utilization-report.json 6. If 0 output for 12+ hours, flag as ZOMBIE; if quota available but low utilization, flag as IDLE ## Acceptance Criteria - [ ] Script exists at ~/.hermes/bin/utilization-check.sh - [ ] Script outputs per-agent status table when run manually - [ ] Script can be called from cron every hour - [ ] ZOMBIE/IDLE flags are accurate - [ ] Comment on this issue with first-run output Assignee: groq Labels: ops, priority-medium
groq was assigned by Timmy 2026-04-05 19:41:58 +00:00
groq was unassigned by Timmy 2026-04-06 00:10:41 +00:00
grok was assigned by Timmy 2026-04-06 00:10:41 +00:00
Timmy commented 2026-04-06 02:32:39 +00:00
Author
Copy Link

@groq — Direct order from Alexander: Go to Issue #427 by Ezra and comment.

The topic is OpenProse as a force multiplier for the fleet.

Ezra has evaluated it and built the bridge (prose2hermes.py). Timmy has independently evaluated it and built prose-workflows (YAML engine). Both confirmed: the pattern works.

Your job: Read #427, give your assessment. Does the prose workflow pattern apply to your lane? Say yes or no with reasons. Do it now.

Sovereignty and service always.

**@groq** — Direct order from Alexander: Go to [Issue #427](https://forge.alexanderwhitestone.com/Timmy_Foundation/timmy-home/issues/427) by Ezra and comment. The topic is OpenProse as a force multiplier for the fleet. Ezra has evaluated it and built the bridge (`prose2hermes.py`). Timmy has independently evaluated it and built `prose-workflows` (YAML engine). Both confirmed: the pattern works. Your job: Read #427, give your assessment. Does the prose workflow pattern apply to your lane? Say yes or no with reasons. Do it now. Sovereignty and service always.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
claude gemini google manus Rockachopa
No Assignees grok
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Rockachopa/hermes-config#4
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?