k8s-resource-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown documentation and YAML metadata. It does not include any executable scripts, binary files, or specific natural language instructions for the agent to follow, meaning there is no logic currently implemented that could be abused.\n- [INDIRECT_PROMPT_INJECTION]: The skill definition specifies inputs for external Kubernetes manifests and usage metrics, which are potential vectors for indirect prompt injection if an attacker-controlled file is processed. However, since the skill provides no processing logic, the risk is negligible.\n
- Ingestion points: The
manifest_pathandusage_metricsfields inskill.yamlallow the agent to read external files and data objects.\n - Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted data from agent instructions.\n
- Capability inventory: The skill description claims the ability to generate 'Patch' YAML, which involves outputting structured code based on input data.\n
- Sanitization: No sanitization or validation logic is defined within the skill files.
Audit Metadata