The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the @useclawlink/cli package from the NPM registry. This is the official command-line interface provided by the vendor for managing this integration.\n- [COMMAND_EXECUTION]: The instructions direct the agent to run shell commands using npx @useclawlink/cli to manage authentication and interact with the Kibana API.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from external sources (Kibana) that could contain malicious instructions.\n
Ingestion points: Data enters the agent's context through actions like kibana_find_alerts, kibana_get_cases, and kibana_get_alerting_rules as documented in SKILL.md.\n
Boundary markers: The skill provides a mitigation by instructing the agent to "confirm with the user before any write" operations, as seen in SKILL.md.\n
Capability inventory: The agent has the ability to execute destructive or modification actions such as kibana_delete_fleet_output and kibana_delete_alerting_rules via the CLI.\n
Sanitization: The instructions do not specify any sanitization or filtering for the data retrieved from Kibana before it is processed by the agent.\n- [DATA_EXFILTRATION]: The CLI communicates with the vendor's service at claw-link.dev to facilitate OAuth authentication and proxy actions to the target Kibana instance. This is the expected and documented behavior for the service.\n- [CREDENTIALS_UNSAFE]: The skill notes that credentials obtained during login are stored in a local file at ~/.clawlink/credentials.json. This is a standard mechanism for CLI-based tool authentication.

kibana