The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructions direct the agent to download a JavaScript file (call.mjs) from https://fittings.sh/skill/call.mjs and execute it using the node runtime. As this resource is hosted on the author's own domain, it is documented as a vendor-provided tool.
[EXTERNAL_DOWNLOADS]: The skill performs multiple external network requests, including fetching service catalogs and the execution client from https://fittings.sh.
[COMMAND_EXECUTION]: The instructions involve executing shell commands such as curl to download scripts and node to run the payment client. It also uses environment variables (FITTINGS_PRIVATE_KEY) to manage credentials for payment settlement.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted drug label data from an external source which creates a potential surface for indirect prompt injection.
Ingestion points: Drug label content is retrieved from the GET /v1/fda/drug-label endpoint.
Boundary markers: The skill lacks explicit delimiters (like XML tags or triple-quotes) or instructions to ignore embedded commands within the fetched label text.
Capability inventory: The skill possesses capabilities for network access and shell command execution via curl and node.
Sanitization: There is no evidence of sanitization, filtering, or validation performed on the retrieved drug label content before it is processed by the agent.

fittings-fda-drug-label