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/prr)<a href="https://agentmods.dev/commands/luanpdd/kit-mcp/prr"><img src="https://agentmods.dev/badge/commands/luanpdd/kit-mcp/prr.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 | $0.00043 | $0.02698 |
| Opus 5 | $0.00022 | $0.01349 |
| Sonnet 5 | $0.00009 | $0.00540 |
| Haiku 4.5 | $0.00004 | $0.00270 |
Grade A, and why
prr 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 — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
6 axes obrigatórios (pular um = aprovação inválida):
- System Architecture — design, dependencies, blast radius, isolation
- Instrumentation/Metrics/Monitoring — 4 golden signals, SLOs, alerting
- Emergency Response — runbooks, on-call, rollback, communication
- Capacity Planning — load testing, scaling, headroom
- Change Management — canary, feature flags, rollback < 60s
- Performance — latency budgets, throughput, optimization
Cria/Atualiza:
.planning/prr/<service>.md(Modo A) OR.planning/prr/feature-<slug>.md(Modo B) — PRR-REPORT.md scored
Após: o user tem decisão Approved / Approved with conditions / Blocked + lista canônica de P0 items por axe + reviewer signature. Phase 40 INT-FW-V2-02 integra /concluir-marco com gate PRR opcional.
Modo A: --service <name> (audit de serviço existente)
Para serviços já em production OU prestes a entrar — agent lê schema (Supabase MCP), Edge Functions code, SLOs definidos (.planning/slos/), advisors. Output: .planning/prr/<service>.md.
Modo B: --feature <description> (audit pré-launch)
Para feature em design/dev — agent lê design docs, SLOs propostos, código WIP. Output: .planning/prr/feature-<slug>.md.
Engagement models (cap 32):
simple— outage cost < $1k/min OR internal tool — 4-8h, 1 sessãoearly— outage cost $1k-100k/min OR customer-facing — semanas, SRE no designplatform— outage cost > $100k/min OR built on Frameworks/SRE Platform — PRR é confirmação
Flags:
--engagement <simple|early|platform>— engagement model (default: AskUserQuestion baseado em outage cost)--reviewer <@handle>— handle do reviewer SRE (default: AskUserQuestion — NUNCA pode ser team dev, anti-pattern auto-PRR)--outage-cost <usd>— custo de outage por minuto (default: AskUserQuestion para escolher engagement)--output <path>— caminho do output (override de default canônico)
Exemplos:
/prr --service orders-api # Modo A — defaults
/prr --service orders-api --engagement early --reviewer @ops-lead # Modo A com config
/prr --feature "RAG sobre documentos privados" --reviewer @sre # Modo B
/prr --service edge-process-emails --engagement simple # Edge Function simples
Pré-requisito (Full mode): projeto Supabase configurado, mcp__supabase__* disponível. Modo offline funciona com fallback graceful (filesystem only — itens MCP-dependentes ficam EVIDENCE_PENDING_MCP).
1. Parsear argumentos (2 modos)
SERVICE=$(echo "$ARGUMENTS" | grep -oE -- '--service [^ ]+' | awk '{print $2}')
FEATURE=$(echo "$ARGUMENTS" | grep -oE -- '--feature "[^"]+"' | sed 's/--feature //; s/^"//; s/"$//')
ENGAGEMENT=$(echo "$ARGUMENTS" | grep -oE -- '--engagement [^ ]+' | awk '{print $2}')
REVIEWER=$(echo "$ARGUMENTS" | grep -oE -- '--reviewer [^ ]+' | awk '{print $2}')
OUTAGE_COST=$(echo "$ARGUMENTS" | grep -oE -- '--outage-cost [^ ]+' | awk '{print $2}')
OUTPUT_PATH=$(echo "$ARGUMENTS" | grep -oE -- '--output [^ ]+' | awk '{print $2}')
# PT-BR: validar mutuamente exclusivos
if [ -n "$SERVICE" ] && [ -n "$FEATURE" ]; then
echo "✗ Erro: --service e --feature são mutuamente exclusivos. Escolha um."
exit 1
fi
# PT-BR: nenhum dos 2 → erro com sugestão
if [ -z "$SERVICE" ] && [ -z "$FEATURE" ]; then
echo "✗ Forneça --service <name> OU --feature \"<descrição>\""
echo " Exemplos:"
echo " /prr --service orders-api"
echo " /prr --feature \"RAG sobre documentos privados\""
exit 1
fi
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 · 206 lines · 43 tokens per session scan A cfa599c1e75d
prr is a command published in the GitHub repository luanpdd/kit-mcp (1 stars, last pushed yesterday), licensed MIT. It adds 43 tokens to every session and 2,698 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.