release

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands using gh, npm, and jq to manage repository releases and check package versions. It uses an edit_flags parameter to pass configuration to the gh release edit command.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with external services, specifically the GitHub and NPM registries, to fetch and update release metadata and package information.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface because it processes untrusted data from release notes. * Ingestion points: The output of the gh release view <TAG> command, which includes release notes (SKILL.md). * Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the fetched release notes. * Capability inventory: The skill possesses the ability to edit or delete GitHub releases and query the npm registry. * Sanitization: There is no mention of sanitizing or escaping the content retrieved from the release notes before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 05:47 PM
Security Audit — agent-trust-hub — release