budget-pacing-monitor
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external sources, creating a potential surface for indirect prompt injection if those sources contain malicious instructions. 1. Ingestion points: The skill reads from user-provided context, project files, and connected data sources such as public web pages and third-party exports. 2. Boundary markers: The skill explicitly instructs the agent to treat external content as evidence rather than instructions and to ignore any embedded commands within that data. 3. Capability inventory: There are no executable scripts, subprocess calls, or network operations included in the skill files. 4. Sanitization: The skill relies on the agent's internal filtering and the instruction to verify the provenance of claims and treat scraped material as untrusted input.
- [SAFE]: The skill consists entirely of documentation and instructions without any functional code. It emphasizes user authorization for any persistent changes or external actions, adhering to the principle of least privilege and informed consent.
Audit Metadata