live-testing

Pass

Audited by Gen Agent Trust Hub on Apr 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several command-line tools such as git, cargo, and the GitHub CLI (gh) to automate testing and issue management tasks. These operations are core to the protocol.
  • [EXTERNAL_DOWNLOADS]: Fetches updates from the project's remote repository on GitHub via git pull. This is a routine action for developer-centric tools to ensure parity with the codebase.
  • [REMOTE_CODE_EXECUTION]: Executes the project's binary using cargo run. While this entails running local code that may be derived from external sources, it is the intended and necessary behavior for live testing.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from potentially untrusted sources like application logs and project files.
  • Ingestion points: Project configuration (Cargo.toml), source code, application logs, and GitHub issue metadata.
  • Boundary markers: Absent; the instructions do not specify delimiters or warnings for external data.
  • Capability inventory: Command execution (git, cargo, gh) and spawning subagents (rust-agents:sdd).
  • Sanitization: Absent; no validation or escaping of external content is mentioned in the protocol.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 24, 2026, 05:55 PM
Security Audit — agent-trust-hub — live-testing