competitive-intelligence
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and synthesize information from external, untrusted sources which could contain malicious instructions.
- Ingestion points: The skill utilizes
web_fetchto retrieve data from external competitor websites, review platforms (G2, TrustRadius, Capterra), and analyst reports inSKILL.md(Step 2: Gather Intelligence). - Boundary markers: While the skill provides a '4-tier source hierarchy' to prioritize information, it does not explicitly instruct the agent to use XML-style delimiters or specific escaping when processing content from these external fetches to prevent instruction hijacking.
- Capability inventory: The skill's primary capabilities are limited to information synthesis and reporting; it does not request tools for file system modification or direct network exfiltration beyond its standard search/fetch operations.
- Sanitization: The skill mitigates risks through professional rigor, requiring verification across multiple sources and the explicit flagging of assumptions (
Assumption:prefix) in the final output.
Audit Metadata