piu-frontend-sync
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Clones frontend repositories from user-provided URLs using
git cloneto perform audits. - [COMMAND_EXECUTION]: Executes shell commands such as
git,mktemp, andrm, and runs a local TypeScript script (bun skills/scripts/piu.ts) to interact with the project and filesystem. - [PROMPT_INJECTION]: Identified a potential surface for indirect prompt injection during the repository scanning process.
- Ingestion points: The skill ingests untrusted code from external repositories (e.g.,
**/api/**,**/services/**) to extract API call patterns (SKILL.md). - Boundary markers: There are no explicit instructions or delimiters used to prevent the agent from following instructions embedded within the scanned source files.
- Capability inventory: The skill has the ability to modify backend project state (creating requests and models) and perform network requests (executing API calls) via the
piu.tsscript (SKILL.md). - Sanitization: The skill lacks documented sanitization or filtering of the content read from external source files before it is interpreted by the agent.
Audit Metadata