The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download a JavaScript client (call.mjs) directly from a remote URL (https://fittings.sh/skill/call.mjs) and execute it using node. While this resource is owned by the skill vendor, downloading and executing code from a remote server at runtime is a significant security risk.
[CREDENTIALS_UNSAFE]: The instructions require users to set a sensitive blockchain private key in the FITTINGS_PRIVATE_KEY environment variable. The skill documentation explicitly acknowledges that the key sits in plaintext wherever the agent runs, which creates a significant risk of credential exposure if the environment is compromised.
[EXTERNAL_DOWNLOADS]: The skill fetches a catalog schema and a execution script from https://fittings.sh. These resources originate from the vendor's infrastructure.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including curl for network requests and node for executing external scripts.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill fetches biography, employment, and publication data from ORCID profiles via the https://fittings.sh/v1/orcid/researcher-profile endpoint.
Boundary markers: None identified; the skill does not appear to use delimiters to separate untrusted external data from the agent's instructions.
Capability inventory: The environment includes shell access (curl, node) and the ability to read environment variables.
Sanitization: There is no evidence of sanitization or validation of the text content retrieved from researcher profiles before it is presented to the agent.

fittings-orcid-researcher-profile