adcp-signals
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent ingests and processes signal metadata (such as names and descriptions) from external signal providers. This data is then used to influence activation decisions, creating a vulnerability surface. 1. Ingestion points: Signal metadata fields like 'name' and 'description' returned by the
get_signalstool. 2. Boundary markers: Absent; the skill instructions do not specify delimiters or instructions to ignore embedded text within the provider metadata. 3. Capability inventory: Theactivate_signaltool allows the agent to execute state-changing operations on external DSP platforms and sales agents. 4. Sanitization: Absent; the documentation does not describe any validation, escaping, or filtering of the metadata before processing. - [SAFE]: No malicious patterns such as command execution, obfuscation, or persistence mechanisms were detected. The skill consists entirely of instructional content and examples for a standard advertising protocol.
Audit Metadata