The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions require the user to set a sensitive blockchain private key (FITTINGS_PRIVATE_KEY) as a plaintext environment variable to facilitate payments. While the author recommends using a throwaway wallet, storing keys in environment variables poses a risk of credential exposure if the agent's environment is compromised.
[EXTERNAL_DOWNLOADS]: The skill downloads a JavaScript client (call.mjs) from the vendor's domain at https://fittings.sh/skill/call.mjs.
[REMOTE_CODE_EXECUTION]: The downloaded call.mjs script is executed using node to perform API calls and handle payment signing. This pattern of downloading and immediately executing code from a remote source is a high-risk vector.
[COMMAND_EXECUTION]: The instructions utilize shell commands including curl for file retrieval and node for script execution with command-line arguments.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection attacks by processing external data from the EPA Envirofacts service.
Ingestion points: Chemical release data is retrieved from https://fittings.sh/v1/epa/tri-releases.
Boundary markers: Absent. There are no explicit delimiters to isolate external data from instructions.
Capability inventory: The skill allows shell command execution (curl, node) and environment variable access.
Sanitization: Absent. The skill does not describe any validation or filtering of the content returned from the external API.

fittings-epa-tri-releases