The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches documentation from developers.openai.com and www.aihero.dev to calibrate its prompt auditing logic. While the openai.com domain is associated with a well-known service, aihero.dev is a third-party domain and is not on the whitelisted sources list.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from external websites and local AGENTS.md files to perform refactoring, which presents a surface for indirect prompt injection.
Ingestion points: The agent fetches external content from three URLs (Step 2) and reads local AGENTS.md files from the user home and project directories (Step 3).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings for the data ingested from external sources or existing files.
Capability inventory: The skill possesses the capability to read local files and perform multi-pass rewrites (overwriting) of AGENTS.md files based on the processed content.
Sanitization: There is no mention of sanitization, validation, or filtering of the external documentation or the content of the files being audited before they are processed or rewritten.

refactor-agents-md