agent-dd
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill operates through a specialized CLI tool,
agent-dd. All shell interactions are restricted to this specific binary as defined in the allowed-tools configuration. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it ingests external data from Datadog (such as logs, trace metadata, and incident descriptions). While this data is untrusted, the skill limits the risk by processing data in structured NDJSON formats and focusing on triage and monitor management rather than general-purpose code execution.
- [CREDENTIALS_UNSAFE]: The skill manages Datadog API and Application keys but explicitly includes safety instructions to prevent credential leakage. It advises against using command-line flags for secrets to avoid exposure in process lists and shell history, recommending secure alternatives like environment variables, piped stdin, or native OS dialogs.
Audit Metadata