{"uid":"cap_0pjzqbB7XI4sPu7vJdiBx","slug":"safe-action-agentutil-net-6d8b98c6","name":"Safe Action Pre-flight Check","description":"Three-part safety pre-flight before destructive or irreversible actions — risk assessment, reversibility check, and timing awareness.","url":"https://safe-action.agentutil.net/v1/skill","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"name":{"enum":["safe-action"],"type":"string","description":"Skill name"}}},"responseSchema":null,"example":{"request":{"name":"safe-action"},"response":{"name":"safe-action","content":"# safe-action\n\nMeasure twice, cut once. Before taking destructive, irreversible, or high-stakes actions, run a three-part safety pre-flight: risk assessment, reversibility check, and timing awareness.\n\nCombines three AgentUtil services — think (safety checklists), undo (reversibility intelligence), and context (situational timing).\n\n## When to Activate\n\nUse before:\n\n- **Deleting** anything — repositories, databases, user accounts, files, DNS records\n- **Deploying** to production — code releases, infrastructure changes, migrations\n- **Mass operations** — bulk updates, batch deletes, sending to many recipients\n- **Permission changes** — revoking access, changing roles, modifying security\n- **Infrastructure changes** — scaling down, terminating instances\n- **Financial actions** — processing refunds, transferring funds\n\n## Workflow\n\n### Step 1: Safety checklist (think)\n```bash\ncurl -X POST https://think.agentutil.net/v1/check \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"action\": \"database_migration\", \"description\": \"Drop column users.email in production\"}'\n```\n\n### Step 2: Reversibility check (undo)\n```bash\ncurl -X POST https://undo.agentutil.net/v1/check \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"platform\": \"GitHub\", \"resource\": \"repository\", \"action\": \"delete\"}'\n```\n\n### Step 3: Timing check (context)\n```bash\ncurl -X POST https://context.agentutil.net/v1/check \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"action\": \"deploy\", \"timezone\": \"America/New_York\", \"platform\": \"github\"}'\n```\n\n### Step 4: Synthesize\n\n| Scenario | Action |\n|----------|--------|\n| All low risk | Proceed. Note key checklist items completed. |\n| Think returns high risk | Pause. Present critical items. Ask for confirmation. |\n| Undo shows irreversible | Warn clearly. Present safer_alternatives. |\n| Context returns high risk | Warn about timing (holiday/after-hours/incident). |\n| Multiple flags | Present all findings. Do NOT proceed without approval. |\n\n## Pricing\n\nEach sub-check: $0.001-$0.003 via x402. Full pre-flight: ~$0.004-$0.008.\nFree exploration: GET /v1/actions (think), GET /v1/platforms (undo), GET /v1/calendar (context).\n\n## Install\n\n- ClawHub: `clawhub install safe-action`\n- MCP: `@agentutil/think-mcp`, `@agentutil/undo-mcp`, `@agentutil/context-mcp`","install":"clawhub install safe-action","service":"https://safe-action.agentutil.net","version":"1.0.1","services":["think.agentutil.net","undo.agentutil.net","context.agentutil.net"],"request_id":"0fdab08a-195f-4bb9-b384-fd803933a64a","description":"Three-part safety pre-flight before destructive or irreversible actions — risk assessment, reversibility check, and timing awareness.","display_name":"Safe Action"}},"exampleRequest":{"name":"safe-action"},"tags":["x402"],"displayCostAmount":"0.001","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"settled","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.001/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.001","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.001","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_VHrPBI9qoNEBPUAwg8h6b","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.001","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Performs a three-part safety pre-flight check (risk assessment, reversibility check, timing awareness) before an AI agent executes a destructive or irreversible action","exampleAgentPrompt":"Before you go ahead and delete those production database records, run the safe-action pre-flight check to assess the risk level, whether it's reversible, and whether now is the right time to do it.","exampleUseCases":null,"resultDescription":"Returns a three-part safety evaluation: a risk assessment indicating the danger level of the proposed action, a reversibility check indicating whether the action can be undone, and timing awareness guidance indicating whether proceeding now is appropriate.","failureModes":["Missing or invalid skill name field returns a validation error","Malformed request body returns 400-level error","Payment not included or insufficient USDC results in 402 Payment Required","Service unavailable returns 5xx with no safety assessment"],"whenToPreferThis":"Use this endpoint whenever an AI agent is about to execute a destructive, irreversible, or high-stakes action and needs a structured safety gate before proceeding. Ideal for agentic workflows where an LLM might autonomously delete files, wipe databases, send bulk communications, or take actions that are hard to undo. Prefer this over ad-hoc LLM reasoning about risk because it provides a consistent, structured three-part framework.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T18:37:21.542Z","isFirstParty":false}