stelow-workflow-entry
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided intent and project names to scaffold a
state.mdfile which influences downstream workflow components. - Ingestion points: User-provided requests for intent classification and the derived project name mentioned in
SKILL.md. - Boundary markers: None specified in the instructions for delimiting user-provided strings within the state file.
- Capability inventory: Shell command execution (
git rev-parse,cp) and local file writing (SKILL.md). - Sanitization: The instructions do not explicitly require the agent to sanitize or validate the short project name or appetite values before interpolating them into the YAML frontmatter of the state file.
- [COMMAND_EXECUTION]: The skill utilizes shell commands to manage the local environment.
- Evidence: Use of
git rev-parse --show-toplevelto determine the project root andcpto bootstrap the state file from a template. - Context: These operations are consistent with the skill's stated purpose of scaffolding a development workflow.
Audit Metadata