The Agent Skills Directory

[SAFE]: No malicious patterns, exfiltration attempts, or unauthorized command executions were detected. The skill follows best practices for process management by requiring user confirmation via ask_user before performing destructive actions.
[COMMAND_EXECUTION]: The skill utilizes the run_shell capability as a fallback mechanism to execute specific ouroboros CLI commands (e.g., ouroboros cancel execution). These commands are restricted to the intended administrative purpose of the skill and do not allow for arbitrary shell command injection.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the ouroboros_project_status tool to identify active executions. While this represents an untrusted data ingestion point, the risk is minimal as the identifiers are used strictly as parameters for cancellation tools, and the instructions include manual verification steps via user prompts.
[DYNAMIC_CONTEXT_INJECTION]: No evidence of dynamic shell-replacement patterns (e.g., exclamation-backtick syntax) was found in the markdown headers or body.

cancel