The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external sources and inject it into the agent's context for "payload enrichment."
Ingestion points: The skill performs a WebFetch on various remote Markdown files hosted on raw.githubusercontent.com as mapped in SKILL.md.
Boundary markers: There are no delimited boundaries or instructions to ignore embedded commands within the fetched content mentioned in the workflow.
Capability inventory: The extracted content is returned directly to the caller to be "baked into executor prompts," potentially influencing downstream agent behavior.
Sanitization: The skill does not implement validation, escaping, or filtering of the content fetched from the remote repository before presenting it to the agent.
[EXTERNAL_DOWNLOADS]: The skill fetches documentation and payload lists from a well-known public repository on GitHub (github.com/swisskyrepo/PayloadsAllTheThings).

patt-fetcher