ci-local
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and executes commands from repository configuration files, creating a vulnerability surface for indirect prompt injection if those files are maliciously modified.\n
- Ingestion points: Reads workflow files from
.github/workflows/*.ymland.github/workflows/*.yaml(SKILL.md, Phase 1).\n - Boundary markers: The skill does not use specific delimiters or instructions to prevent the execution of embedded malicious instructions within the workflow files.\n
- Capability inventory: Utilizes the
Bashtool to execute translated shell commands (SKILL.md, Phase 3).\n - Sanitization: No sanitization, validation, or filtering is applied to the command strings extracted from
run:blocks before they are executed.\n- [COMMAND_EXECUTION]: The skill is designed to translate and execute arbitrary shell commands parsed from local repository files using theBashtool (SKILL.md, Phase 3).
Audit Metadata