retry-utils
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of arbitrary shell commands through the {baseDir}/retry.sh script. This creates a high-privilege execution interface for the agent to run system-level operations.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. If the agent incorporates untrusted data into the --command, --error-pattern, or --until-pattern arguments without sanitization, it could lead to command injection or logic bypass. • Ingestion points: Command line arguments for --command, --error-pattern, and --until-pattern in SKILL.md. • Boundary markers: No delimiters or protective markers are provided to isolate untrusted data from instructions. • Capability inventory: General shell command execution through the retry.sh script. • Sanitization: No sanitization, validation, or escaping logic is described for the input strings.
- [NO_CODE]: The referenced script retry.sh is not included in the provided skill files, making it impossible to verify the security of the underlying execution logic.
Audit Metadata