intent-signal-aggregator-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from various external web sources, such as job postings, news articles, and social media, to identify buyer intent signals. This content is ingested into the agent's context without sanitization, creating a surface for indirect prompt injection attacks.
- Ingestion points: External job listings (e.g., LinkedIn Jobs), funding news (e.g., TechCrunch), and leadership change announcements in SKILL.md.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the workflow to separate untrusted data from the agent's core logic.
- Capability inventory: The skill utilizes HubSpot search tools (
mcp__claude_ai_Epiphan_Ai__hubspot_search_companies,mcp__claude_ai_Epiphan_Ai__hubspot_search_contacts) and has the ability to write JSON metrics to the local file system at~/.claude/skill-analytics/. - Sanitization: No sanitization, escaping, or validation of the external content is defined before the data is interpolated into the prompt or used for scoring.
Audit Metadata