The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted data (Terms of Service and Privacy Policies) provided by the user. This creates a surface for indirect prompt injection if the analyzed text contains malicious instructions.
Ingestion points: User-provided document text or excerpts analyzed via the provided methodology.
Boundary markers: While the skill uses markdown headers and examples to structure output, it lacks explicit "ignore instructions" delimiters to protect the agent from adversarial content within the documents being analyzed.
Capability inventory: The skill focuses on text analysis and categorization; however, it references a custom MCP server that may provide additional capabilities.
Sanitization: No input validation or sanitization routines are defined in the instructions.
[EXTERNAL_DOWNLOADS]: The skill specifies a dependency on the @clawfu/mcp-skills Node.js package in its YAML frontmatter to provide the necessary MCP server environment.

terms-analyzer