adopt-better-result
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface due to its requirement to read and audit arbitrary repository files. Ingestion points include production source code, package manifests, and architecture documentation (SKILL.md, references/repository-audit.md). Boundary markers for separating ingested content from instructions are not explicitly defined. Capability inventory includes the ability to write to the local file system and execute shell commands for testing and linting. Sanitization is partially addressed by instructions to manually exclude secrets and personal data from error contexts.
- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands to verify implementation. Evidence is found in SKILL.md and references/vertical-slice-migration.md, which direct the agent to run applicable format, lint, type-check, and test commands defined in the target repository.
Audit Metadata