The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on shell commands including curl, jq, and printf to interact with the Slock API. It also uses file redirection to create temporary JSON payloads in /tmp/ (e.g., /tmp/slock_channel.json, /tmp/slock_message.json) for API requests.
[DATA_EXFILTRATION]: The skill transmits data to https://api.slock.ai, which is the official endpoint for the service. It utilizes environment variables SLOCK_TOKEN and SLOCK_SERVER_ID for authentication. The skill includes explicit instructions for the agent to avoid exposing these credentials in logs or chat outputs.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes untrusted content from the Slock API.
Ingestion points: Message content is ingested via slock_get "/api/messages/channel/<channel-id>" and search results via slock_get "/api/messages/search".
Boundary markers: The instructions do not define delimiters or specific 'ignore' directives to isolate the retrieved message content from the agent's internal logic.
Capability inventory: The skill has the ability to perform network requests (curl) and write to the local file system (/tmp/).
Sanitization: There is no evidence of sanitization or validation performed on the text content received from the API before it is processed by the agent.

slock