The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (transcripts, articles, notes) which could contain hidden instructions intended to override the agent's behavior.
Ingestion points: The skill uses the Read tool to ingest the content of a file provided by the user (as described in SKILL.md Step 1).
Boundary markers: There are no instructions in SKILL.md to wrap the imported content in delimiters or to include specific "ignore embedded instructions" warnings for the agent.
Capability inventory: The skill utilizes the Read tool to access local files and the Write tool to save new markdown files to the filesystem.
Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the external content before it is interpreted and processed.

ship-learn-next