The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from the Nestr platform which can be modified by other users, creating a surface for instructions that could influence the agent's behavior.
Ingestion points: Content is retrieved through nestr search, nestr nests get, nestr comments list, and nestr nests activity (SKILL.md).
Boundary markers: No specific delimiters or "ignore instructions" markers are provided to separate data from system instructions.
Capability inventory: The skill allows for creating and updating nests, uploading files, and posting comments or direct messages across the platform (SKILL.md).
Sanitization: While the server performs HTML sanitization for UI rendering, no evidence of content validation for the agent's context is present.
[COMMAND_EXECUTION]: The skill relies on the nestr CLI to perform operations, including file system interactions.
The nestr nests files read ... --save <path> command allows the agent to download files from the platform to the local disk, which could potentially overwrite existing local files if the path is not validated.
The nestr nests files upload command permits the agent to read local files and upload them to the Nestr platform.

nestr-basics