About loop-budget
loop-budget is an agent skill for checking an automated loop's token spend, daily caps, and pause flags at the start and end of each run. It throttles or exits when limits are reached and records run statistics.
Key features
- 01Daily cap and pause-flag checks
- 02Token usage tracking by pattern
- 03Report-only mode at budget threshold
- 04Early exit when no work is actionable
- 05Run-log and alert updates
Use cases
- 01Limiting token use in recurring automation
- 02Stopping a loop when its budget is exhausted
- 03Avoiding agent spawns when nothing needs action
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ cobusgreyling/ loop-engineering --skill loop-budgetAfter installing, type/loop-budgetin Claude Code to confirm it loaded.
Details
From the project READMELoop Budget Guard
Run at the start and end of every loop iteration.
Start of run
- Read
loop-budget.mdfor daily caps and kill-switch flags. - Read recent entries in
loop-run-log.md(last 24h). - Sum
tokens_estimatefor the active pattern today. - If spend ≥ 80% of the pattern's daily cap → report-only mode (no sub-agents, no auto-fix).
- If spend ≥ 90% and High Priority items remain in
STATE.md, yield to the budget-negotiator skill (if installed). Otherwise, if spend ≥ 100% orloop-pause-allis set → exit immediately with a one-line note in STATE.md. - If watchlist/state has no actionable items → exit in <5k tokens (do not spawn sub-agents).
End of run
Append one JSON object to loop-run-log.md:
{
"run_id": "<ISO8601>",
"pattern": "<pattern-id>",
"duration_s": <number>,
"items_found": <number>,
"actions_taken": <number>,
"escalations": <number>,
"tokens_estimate": <number>,
"outcome": "no-op | report-only | fix-proposed | escalated"
}
Rules
- Never exceed
max sub-agent spawns/runfromloop-budget.md. - High-cadence patterns (CI Sweeper, PR Babysitter) must early-exit when nothing is actionable.
- On self-throttle, append a line to
loop-budget.mdunder Alerts This Period.
Comments
Markdown supported