create-template

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill generates YAML templates that accept runtime inputs via <+input> placeholders, specifically within shell command blocks (e.g., command: <+input>). This creates a vulnerability surface where malicious input could lead to command injection in the CI/CD environment.
  • Ingestion points: Runtime input values replace the <+input> placeholders in generated Harness YAML templates (found in SKILL.md).
  • Boundary markers: There are no explicit instructions or generated delimiters to sanitize or wrap user-provided inputs to prevent shell command injection.
  • Capability inventory: The resulting templates are pushed via the harness_create tool and eventually executed by Harness CI/CD runners, which have the capability to execute shell commands, perform file system operations, and make network requests.
  • Sanitization: The skill lacks instructions for validating or escaping runtime inputs before they are interpolated into the shell command strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 11:24 PM
Security Audit — agent-trust-hub — create-template