How this skill is triggered — by the user, by Claude, or both
Slash command
/honeycomb:sloThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Dependency chain: derived column (SLI) → SLO → burn alert. All commands require `--dataset`.
Dependency chain: derived column (SLI) → SLO → burn alert. All commands require --dataset.
target_per_million: 99.9% = 999000, 99% = 990000, 95% = 950000.
slo create -f - (file-only, no flags yet — see #109)slo update <id> --name/--description/--target/--time-period (flag-based, read-modify-write) or -f (mutually exclusive)slo get <id> / slo get <id> --detailed (Enterprise-only)slo list / slo delete <id> --yesCreate body: {"name":"...","sli":{"alias":"<derived-column-alias>"},"time_period_days":30,"target_per_million":999000}
Create/update require recipients (non-empty array of {"id":"..."}) and slo as {"id":"..."} (nested, not flat slo_id).
Exhaustion time: {"alert_type":"exhaustion_time","exhaustion_minutes":240,"slo":{"id":"..."},"recipients":[{"id":"..."}]}
Budget rate: {"alert_type":"budget_rate","budget_rate_window_minutes":60,"budget_rate_decrease_threshold_per_million":50000,"slo":{"id":"..."},"recipients":[{"id":"..."}]}
All burn alert commands are file-only (-f), no flags yet (#110, #111).
slo history --slo-id <id> --start-time <unix> --end-time <unix> — --slo-id is repeatable. Response is a map keyed by SLO ID.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub bendrucker/claude --plugin honeycomb