The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources such as G2 reviews, Capterra reviews, and LinkedIn profiles.
Ingestion points: Data is ingested through tools like mcp__coldiq__find_signals or via scraping platforms like Clay and Phantombuster as described in the fallback plays in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are specified for the processed reviews or profile data.
Capability inventory: The skill uses mcp__coldiq__* tools for signal finding, person search, and enrichment, and provides templates for automated outreach.
Sanitization: The instructions do not include steps to sanitize or validate the content retrieved from external review sites or social media profiles before interpolation into messages or further processing.
[DYNAMIC_EXECUTION]: The skill instructs users to install and run the vendor's tool using a package runner.
Evidence: The instruction npx -y @coldiq/mcp@latest in SKILL.md dynamically downloads and executes the latest version of the package from the NPM registry.
[EXTERNAL_DOWNLOADS]: The skill communicates with external vendor infrastructure and public package registries.
Evidence: It utilizes the API at api.cold-iq.com and directs users to download the @coldiq/mcp package from the NPM registry.

competitor-signals