meta-ads-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a pure analysis extension. It contains no executable scripts, binaries, or commands. It is restricted to reading data and providing text-based interpretation, posing no direct threat to the host system or user data.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it is designed to ingest and process untrusted external data from Meta Ads exports.
- Ingestion points: The skill accepts CSV files, pasted tables, and API data as primary inputs (SKILL.md, Phase 0).
- Boundary markers: Absent. The skill does not instruct the agent to use specific delimiters or to ignore potential instructions embedded within the ad data.
- Capability inventory: Analysis is restricted to text generation. The skill lacks capabilities for file system modification, network exfiltration, or subprocess execution, significantly mitigating the risk of exploitation.
- Sanitization: Absent. There is no explicit validation or filtering of the ingested data content.
Audit Metadata