Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/luanpdd/kit-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/luanpdd/kit-mcp/burn-rate-status)<a href="https://agentmods.dev/commands/luanpdd/kit-mcp/burn-rate-status"><img src="https://agentmods.dev/badge/commands/luanpdd/kit-mcp/burn-rate-status.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00031 | $0.05742 |
| Opus 5 | $0.00015 | $0.02871 |
| Sonnet 5 | $0.00006 | $0.01148 |
| Haiku 4.5 | $0.00003 | $0.00574 |
Grade A, and why
burn-rate-status scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 409 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lê: .planning/slos/*.yml (definição com alert_thresholds.page + .ticket) + .planning/metrics/snapshots/*.json (eventos persistidos via metrics.persistSnapshot() — Phase 99 + Phase 102 auto-snapshot).
Cria/Atualiza: nada — comando read-only.
Após: o user vê tabela com colunas fast_burn, slow_burn, combined (status PAGE / TICKET / WARN / OK) e pode escolher invocar /investigar-producao se há burn ativo, ou aguardar mais snapshots se ambas janelas estão no_data.
Flags (defaults dual-window — Phase 103):
--fast-baseline <duration>— janela fast (page-tier). Default:1h.--slow-baseline <duration>— janela slow (ticket-tier). Default:6h.--format <table|json>— output format. Default:table.
Combinações canônicas (skill burn-rate-alerting):
- Fast (page-tier): lookahead 1h, baseline 5m, multiplier 14.4× — esgota ~2% do budget mensal em 1h.
- Slow (ticket-tier): lookahead 6h, baseline 30m, multiplier 6× — esgota ~10% do budget mensal em 6h.
Fator 4×: lookahead ≤ 4× baseline para extrapolação confiável (skill rule). 1h ≤ 4× 15m e 6h ≤ 4× 90m são ambos respeitados; defaults operacionais são page (1h baseline) + ticket (6h baseline) — a janela lookahead propriamente dita está embutida nos alert_thresholds.{page,ticket}.lookahead do YAML do SLO.
Phase 99 + 102 wiring: este comando consome dados persistidos automaticamente pelo handler MCP metrics-snapshot (Phase 102 OBS-20-01 — auto-persist via persistSnapshot() em cada call com throttle 1s). Sem snapshots na janela, o comando emite "no_data" para o SLO em vez de inventar números.
Cross-reference: este comando é a implementação do pattern "dashboard de burn rate" canônico documentado em kit/skills/burn-rate-alerting/SKILL.md. A skill é a SSOT da fórmula e dos thresholds; este comando é o renderer.
1. Parsear argumentos
Bash:
SLO_NAME=$(echo "$ARGUMENTS" | awk '{for(i=1;i<=NF;i++) if($i !~ /^--/) {print $i; exit}}')
FAST_BASELINE=$(echo "$ARGUMENTS" | grep -oE -- '--fast-baseline [^ ]+' | awk '{print $2}')
SLOW_BASELINE=$(echo "$ARGUMENTS" | grep -oE -- '--slow-baseline [^ ]+' | awk '{print $2}')
FORMAT=$(echo "$ARGUMENTS" | grep -oE -- '--format [^ ]+' | awk '{print $2}')
[ -z "$FAST_BASELINE" ] && FAST_BASELINE="1h"
[ -z "$SLOW_BASELINE" ] && SLOW_BASELINE="6h"
[ -z "$FORMAT" ] && FORMAT="table"
Convert duration to ms (helper):
to_ms() {
local d="$1"
case "$d" in
*h) echo $(( ${d%h} * 3600000 ));;
*m) echo $(( ${d%m} * 60000 ));;
*s) echo $(( ${d%s} * 1000 ));;
*d) echo $(( ${d%d} * 86400000 ));;
*) echo 0 ;;
esac
}
FAST_BASELINE_MS=$(to_ms "$FAST_BASELINE")
SLOW_BASELINE_MS=$(to_ms "$SLOW_BASELINE")
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 409 lines · 31 tokens per session scan A 46aad6cb31d6
burn-rate-status is a command published in the GitHub repository luanpdd/kit-mcp (1 stars, last pushed today), licensed MIT. It adds 31 tokens to every session and 5,742 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.