strategy
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection due to its data ingestion patterns. It retrieves untrusted content from external sources (LinkedIn posts via Apify and website data via scrapers) and uses this data to generate strategies and populate Notion databases. It lacks explicit boundary markers (e.g., delimiters) or instructions to treat the ingested data as non-executable text. A malicious LinkedIn profile containing hidden instructions could potentially influence the agent's behavior during the strategy creation process.
- Ingestion points: LinkedIn profile posts (via apimaestro/linkedin-profile-posts) and website content (via website-scraper).
- Boundary markers: Not present in the instructions.
- Capability inventory: Writing and updating pages in Notion, accessing GTM-strategy plugin data.
- Sanitization: No evidence of sanitization or filtering of the scraped content.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the retrieval of information from arbitrary external URLs provided in the arguments and interacts with several third-party tools (Apify actors, website scrapers). While these operations are central to the skill's purpose, they involve processing remote content from sources outside the immediate control of the user or the agent.
Audit Metadata