القائمة
Tracks cumulative API spend against a monthly budget and pauses non-essential automations when thresholds are crossed
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | spend-circuit-breaker |
| version | 1.0 |
| category | openclaw-native |
| description | Tracks cumulative API spend against a monthly budget and pauses non-essential automations when thresholds are crossed |
| stateful | true |
| cron | 0 */4 * * * |
OpenClaw has no built-in hard spending cap. This skill monitors session logs to estimate cumulative API cost, alerts at configurable thresholds, and automatically pauses non-essential cron automations when the monthly budget ceiling is hit.
Before the skill is useful, record your monthly budget in state:
python3 check.py --set-budget 50 # $50/month hard cap
python3 check.py --set-alert 0.5 0.75 # Alert at 50% and 75%
Runs every 4 hours (cron: "0 */4 * * *"). On each wakeup:
~/.openclaw/skill-state/spend-circuit-breaker/state.yaml — load monthly_budget_usd and spend_this_month_usd~/.openclaw/sessions/ to estimate new spend since last_checked_at (model × token counts × price table)spend_this_month_usd and last_checked_at in state| Spend % | Action |
|---|---|
| < 50% | Log silently — nothing to surface |
| ≥ 50% | Notify user: "You've used ~50% of your $X budget this month" |
| ≥ 75% | Notify user + suggest which cron skills to pause |
| ≥ 100% | Notify user + automatically pause all cron-scheduled skills except daily-review, morning-briefing, and spend-circuit-breaker itself |
At 100%, write circuit_open: true to state. install.sh checks this flag before registering new cron jobs — new installs are allowed, new cron triggers are not.
To restore automations after a budget reset or manual override:
python3 check.py --reset-circuit
This sets circuit_open: false and re-registers all paused cron jobs.
The script ships with a price table for common models (claude-3-5-sonnet, gpt-4o, etc.). To add a custom model:
python3 check.py --add-model my-model-name --input-cost 3.00 --output-cost 15.00
Prices are per million tokens.
Handles PR review feedback by fetching comments, grouping issues, fixing one group at a time, and verifying before replies.
Reviews whether a skill will trigger reliably, guide useful behavior, avoid overlap, and produce testable outcomes.
Diagnoses OpenClaw provider, fallback, channel, MCP, and gateway config issues with read-only scans and stateful summaries.
Tracks required validation gates, records pass/fail/waived results, and reports readiness before task completion.
YAML-based delegation grant ledger — issues, validates, and tracks scoped permission grants for sub-agent expansions with token budgets and auto-expiry.
Walks the memory DAG to recall detailed context on demand — query, expand, and assemble cited answers from hierarchical summaries without re-reading raw transcripts.