The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from external job boards at api.ashbyhq.com and provides tools to execute tasks or browse the site (unbrowse.run, unbrowse.browse.act). This creates an attack surface where malicious instructions embedded in the processed job board data could potentially influence the agent's behavior.
Ingestion points: External data enters the agent context via the api_ashbyhq_com__get_job_board_notion tool and generic unbrowse.run/unbrowse.scrape calls in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands were found in the skill configuration.
Capability inventory: The skill includes capabilities to perform tasks, scrape pages, and interact with the browser (unbrowse.run, unbrowse.scrape, unbrowse.browse.act) as described in SKILL.md.
Sanitization: There is no evidence of sanitization or filtering applied to the external content before it is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill configuration and documentation reference external resources for installation and runtime execution.
The installation instructions in README.md and SKILL.md suggest using npx skills add pointing to the vendor's infrastructure.
The server.json and SKILL.md files configure a remote MCP server at https://unbrowse.ai/mcp/ashbyhq.com.

ashbyhq