The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from external URLs and local files, creating a potential vector for indirect prompt injection attacks.
Ingestion points: Reads content from live URLs, dist/index.html, src/pages/, and documentation files (POSITIONING.md, CONTENT_GUIDE.md, STORY.md) found in SKILL.md.
Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent ignores malicious instructions embedded within the site content.
Capability inventory: The skill utilizes shell commands (npm run build, git status, git check-ignore) and performs file writes to the analytics/ directory as described in the Handoff section.
Sanitization: The instructions lack specific sanitization or validation steps for the content being processed.
[COMMAND_EXECUTION]: The skill directs the agent to execute local build scripts, which can lead to arbitrary code execution.
Evidence: The skill instructs the agent to run npm run build in the Inspect section to render the homepage for analysis. This executes scripts defined in the local environment's package.json file.

website-positioning-check