The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to install software and perform browser-based research.
Evidence: npm install -g agent-browser && npx playwright install chromium in SKILL.md is used for environment setup.
Evidence: The skill uses CLI commands like agent-browser screenshot and agent-browser get text body to interact with external websites.
[EXTERNAL_DOWNLOADS]: The skill downloads and installs third-party packages from public registries.
Evidence: Installation of agent-browser from the npm registry.
Evidence: Downloads the Chromium browser via Playwright's official installation command.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its broad ingestion of untrusted web content combined with codebase-modification capabilities.
Ingestion points: The skill fetches the full text body of external websites (competitor pricing pages, G2 reviews, Reddit, and the Wayback Machine) as defined in the 'Research Mode' section of SKILL.md.
Boundary markers: Absent. The agent is instructed to extract information from the raw text body without the use of delimiters or 'ignore' instructions for embedded content.
Capability inventory: The skill features an 'Implementation Mode' (SKILL.md Context B) which allows the agent to modify the local codebase, including templates, routing, and dependencies. It also has command execution capabilities for package installation.
Sanitization: No sanitization or validation of the fetched web content is performed before it is processed by the agent to generate recommendations or code changes.

marketing-pricing