The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the Bash tool to execute a wide variety of diagnostic commands, including system resource checks (top, df, lsof), container orchestration commands (kubectl), and database queries (psql, mysql, redis-cli). While these are aligned with the skill's purpose, they provide a broad execution surface.
[DATA_EXFILTRATION]: The skill is instructed to read sensitive information such as application logs, environment variables (.env changes), and configuration files that may contain credentials or PII. When combined with the WebFetch and WebSearch tools, this creates a potential risk for data exfiltration if the agent's logic is subverted.
[PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection due to the nature of the data it processes. Ingestion points: The agent reads application logs, web server logs, container events, git commit messages, and deployment manifests, all of which can contain attacker-influenced data. Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the ingested data. Capability inventory: The skill possesses powerful capabilities including Bash execution, file writing (Write, Edit), and network access (WebFetch). Sanitization: There is no mention of sanitizing or validating the content of logs or external data before the agent processes it, allowing potentially malicious instructions in logs to influence agent behavior.

incident-responder