competitive-strategy-tracker
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow for scanning and analyzing competitor landscape data from external sources including website taglines, hero sections, blog posts, and social media content (Step 1, Dimensions A-E). This ingestion of untrusted data from the open web creates a potential vulnerability where malicious instructions embedded in a competitor's site could influence the agent's reasoning.
- Ingestion points: Competitor websites (hero sections, pricing pages), social media posts (LinkedIn, Twitter), and review sites (G2, Capterra) as specified in the scanning dimension sections.
- Boundary markers: The instructions lack explicit boundary markers or delimiters to isolate the scraped external content from the agent's core strategic reasoning instructions.
- Capability inventory: The skill writes analysis results to the local filesystem (clients//competitors/) and passes strategic recommendations to downstream skills like
cold-email-outreachandemail-drafting. - Sanitization: There are no mentioned processes for sanitizing, escaping, or filtering the extracted competitor data before it is processed by the model for strategy and outreach generation.
- [COMMAND_EXECUTION]: The skill metadata includes a standard installation command using
npx. This is an expected pattern for skill distribution on the platform and does not indicate malicious intent, but it involves the execution of a package manager command.
Audit Metadata