seasonal-aso
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill functions as an instructional guide for marketing tasks. It does not contain any obfuscated code, unauthorized command execution, persistence mechanisms, or credential harvesting patterns. The shell command examples provided in the documentation are for API interaction illustrative purposes and do not trigger dangerous subprocesses.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data by instructing the agent to read external context and API data.
- Ingestion points: The skill reads a local file named
app-marketing-context.mdand fetches trending keyword data from the author's external API endpoints (/v1/keywords/trending, etc.). - Boundary markers: None are defined in the skill to distinguish between instructions and ingested data.
- Capability inventory: The skill is limited to text generation and metadata recommendations. It does not perform file writes, network exfiltration of sensitive data, or shell command execution beyond the context of the user-facing recommendations.
- Sanitization: No sanitization or validation of the external API data or local file content is implemented.
- Despite the ingestion surface, the lack of dangerous capabilities makes this a minimal risk.
Audit Metadata