The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and analyzes content from external, untrusted sources such as live website URLs for audits and competitor research.
Ingestion points: Data enters the agent context through the agent-browser and WebFetch tools which are used to audit live URLs and perform competitor research as described in shared-patterns.md.
Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted web content from the agent's core instructions.
Capability inventory: The skill possesses the ability to write files to the local file system (deliverables saved in .pawbytes/ as defined in deliverables.md), execute shell scripts (tool-discovery.sh in shared-patterns.md), and perform browser automation.
Sanitization: There is no evidence of sanitization or filtering applied to external content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill provides instructions in shared-patterns.md to download and install the agent-browser tool from Vercel Labs' official GitHub repository (https://github.com/vercel-labs/agent-browser) using npx or npm. These references target a well-known and trusted organization.
[COMMAND_EXECUTION]: The skill instructs the agent to run local shell and batch scripts for environment setup and tool discovery (tool-discovery.sh and tool-discovery.bat) as part of the 'Pre-Flight' routine defined in shared-patterns.md. These scripts are part of the broader marketing suite from the same vendor.

paw-mkt-video