track-pipeline
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes job application data (company names, roles, notes) from
tracker.mdortracker.csv, which serves as an ingestion point for potentially untrusted data if populated by other web-scraping skills. - Ingestion points: Reads application data from
tracker.mdandtracker.csv. - Boundary markers: Data is structured within markdown/CSV tables, but there are no explicit instructions to the agent to ignore embedded commands within those fields.
- Capability inventory: The skill instructions are limited to file read/write operations; no shell execution, network capabilities, or dangerous subprocess tools are defined.
- Sanitization: There is no explicit sanitization or validation of the content within the job tracker fields.
- [NO_CODE]: The skill consists entirely of markdown instructions and a template file. It does not include any scripts (Python, JavaScript, Bash) or compiled binaries, significantly reducing the attack surface.
Audit Metadata