debug-pipeline
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and analyzes external content from pipeline execution logs and YAML definitions, creating a vulnerability surface for indirect prompt injection attacks where instructions hidden in logs could attempt to influence agent behavior.
- Ingestion points: The skill uses
harness_diagnoseandharness_gettools inSKILL.mdto fetch diagnostic data and logs directly from the Harness environment into the agent's context. - Boundary markers: There are no explicit instructions or delimiters defined to isolate the untrusted log content from the agent's own instructions, potentially allowing the agent to misinterpret data as commands.
- Capability inventory: The skill provides extensive access to the Harness platform through diagnostic tools, including the ability to list executions, get detailed resource information, and retrieve logs.
- Sanitization: The instructions do not specify any sanitization, filtering, or validation steps for the retrieved logs before they are processed for root cause analysis.
Audit Metadata