rhdh-release-sos

Warn

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The script scripts/_release_bridge.py dynamically calculates a file path to a sibling skill directory (rhdh-release-status/scripts) and appends it to sys.path to import internal modules. This pattern introduces a runtime dependency on code loaded from a computed path outside the skill's package.
  • [COMMAND_EXECUTION]: The skill executes system commands via uv run and calls the acli Jira CLI to perform search queries and retrieve release milestone dates.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Jira issue summaries and descriptions to build the Scrum of Scrums report.
  • Ingestion points: Jira issue data fetched via acli search results.
  • Boundary markers: No specific delimiters or warnings are used when interpolating Jira content into report templates.
  • Capability inventory: File system access for writing reports and shell command execution for Jira queries.
  • Sanitization: The skill performs HTML escaping for its HTML report output, though the markdown report uses direct string interpolation.
  • [EXTERNAL_DOWNLOADS]: The skill generates links to redhat.atlassian.net for issue tracking. This is a well-known service associated with the developer's infrastructure.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 23, 2026, 06:18 PM