integrate-delivery

Warn

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill executes a Node.js script using a dynamically resolved path (node /scripts/deliver-task.mjs). This pattern relies on the agent correctly identifying the peer skill's location, which introduces risk if path resolution is manipulated.\n- [COMMAND_EXECUTION]: During the 'integration validation' phase, the skill executes 'destination project instructions'. This involves running arbitrary commands defined within the code repository being integrated, which is a standard execution surface for untrusted project code.\n- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through external data ingestion.\n
  • Ingestion points: Reads delivery.json, commit-range metadata, and project-specific instructions from the target Git repository.\n
  • Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands in the ingested data.\n
  • Capability inventory: The skill has the ability to execute shell commands (Git), run Node.js scripts, and execute arbitrary project validation commands.\n
  • Sanitization: No evidence of sanitization or filtering for the data read from the repository or delivery artifacts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 31, 2026, 05:42 AM
Security Audit — agent-trust-hub — integrate-delivery