dx-roaster

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use shell commands like git clone and gh api to retrieve repository contents and metadata. These commands utilize user-provided repository URLs as input.
  • [EXTERNAL_DOWNLOADS]: The skill performs network operations to download content from GitHub. These downloads target a well-known service and are necessary for the skill's stated functionality.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and analyzes untrusted content from the repositories it audits.
  • Ingestion points: The agent reads the full content of README.md, CONTRIBUTING.md, and other project files from the target repository (SKILL.md).
  • Boundary markers: There are no instructions specifying the use of delimiters or 'ignore' warnings when the agent processes this external content.
  • Capability inventory: The skill has the capability to execute shell commands (git, gh) and write output files to the local filesystem (SKILL.md).
  • Sanitization: The skill lacks any defined validation or sanitization processes for the data ingested from scanned repositories.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 10:41 PM
Security Audit — agent-trust-hub — dx-roaster