The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by processing external audit files which may contain content derived from public websites.
Ingestion points: The workflow involves reading and parsing markdown files located in ~/.geo-prospects/audits/ to extract scores, crawler statuses, and action items.
Boundary markers: The skill lacks explicit instructions or delimiters to isolate data being parsed from the audit files, which could lead the agent to interpret embedded instructions as its own.
Capability inventory: The skill utilizes Bash, Read, and Write tools to manage files and generate reports, providing a functional path for potentially injected instructions.
Sanitization: There is no evidence of content sanitization or validation performed on the text extracted from the audit files before it is interpolated into the final report or used for logic branching.

geo-compare