apollo-incident-runbook

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches status information from the official Apollo.io status page at status.apollo.io and interacts with the api.apollo.io endpoints. These are well-known, trusted sources for the service being managed.
  • [COMMAND_EXECUTION]: Utilizes kubectl to perform incident mitigation tasks such as scaling deployments, updating environment variables, and restarting services. These actions align with the skill's purpose as an incident response runbook.
  • [REMOTE_CODE_EXECUTION]: A diagnostic script uses python3 -c for parsing JSON data retrieved from a trusted status API. This is a common pattern for processing structured output in shell environments.
  • [PROMPT_INJECTION]: The skill ingests data from an external status API, creating a potential surface for indirect prompt injection.
  • Ingestion points: Data is fetched from status.apollo.io/api/v2/status.json in SKILL.md.
  • Boundary markers: No delimiters or instructions are used to separate ingested data.
  • Capability inventory: Uses kubectl, curl, and file writing.
  • Sanitization: Ingested data is processed via python3 and jq without explicit sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 11:17 PM
Security Audit — agent-trust-hub — apollo-incident-runbook