The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes curl to interact with the vendor's CRM API at crm.psquared.dev. These commands are used to query for existing records, create new company and person entries, and document lead justifications via GraphQL mutations.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites to perform lead validation.
Ingestion points: Untrusted data enters the agent context via the WebFetch tool when visiting potential lead websites (described in SKILL.md).
Boundary markers: There are no explicit instructions for using delimiters or boundary markers when the agent processes the fetched website content.
Capability inventory: The skill possesses file-read capabilities (reading .env and skip-list.json) and network-write capabilities via curl to the vendor CRM.
Sanitization: The instructions do not specify automated sanitization or escaping of the content retrieved from external websites before it is used to generate justifications or CRM notes.
[CREDENTIALS_SAFE]: The skill follows standard security practices by instructing the agent to read API credentials (PSQUARED_CRM_TOKEN) from a local .env file rather than hardcoding them in the instructions.

find-leads