release-manager

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it is designed to process untrusted external data.
  • Ingestion points: The skill explicitly mentions processing "Web content and external data (e.g., via BrowserOS)" in SKILL.md.
  • Boundary markers: The protocol includes explicit instructions to treat external data as "hostile" and requires the use of a sequentialthinking loop to assess risk before tool execution.
  • Capability inventory: The agent is granted terminal access via the ExecutionProxy Interface (using the rtk prefix) and the ability to write persistent artifacts via the MemoryStore Interface (targeting docs/departments/).
  • Sanitization: Instructions mandate the redaction of secrets and PII (Personally Identifiable Information) before sharing context with subagents.
  • [COMMAND_EXECUTION]: The skill references the ExecutionProxy Interface for running terminal commands (e.g., rtk npm test). While this is presented as a controlled execution path within the framework's protocols to minimize overhead, it confirms the agent's capability to execute shell commands within the project environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 06:16 AM
Security Audit — agent-trust-hub — release-manager