The Agent Skills Directory

[SAFE]: No malicious behavior, obfuscation, or safety bypass attempts were detected. The skill follows established patterns for market research and data organization.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from the web via Perplexity search results in Steps 2, 3, and 4. This data is used to populate competitive analysis reports but does not interact with high-privilege system components or executable shells. * Ingestion points: Perplexity reasoning and factual search tools (Step 2, 3, 4). * Boundary markers: Findings are constrained to the structured markdown schema defined in references/competitive-intel-schema.md. * Capability inventory: Limited to reading and writing markdown files within research-memory/ and brand-memory/ directories. * Sanitization: External search content is directly interpolated into markdown fields without explicit sanitization.

competitor-finder