perses-dashboard-create
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute 'percli' commands for linting, applying, and describing dashboards. These operations are restricted to the management of Perses resources and are consistent with the skill's stated purpose.
- [DATA_EXFILTRATION]: While 'WebFetch' and 'WebSearch' tools are enabled, the skill instructions do not contain any patterns for sending sensitive information to external servers. The network interaction is limited to the Perses server via 'percli' or MCP tools for legitimate deployment tasks.
- [PROMPT_INJECTION]: The skill instructions are focused on requirement gathering and dashboard generation. There are no attempts to override agent safety guidelines or hide malicious instructions in the prompts.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided requirements to generate dashboard definitions. It mitigates potential risks by enforcing a validation phase ('Phase 3: VALIDATE') where the generated configuration is checked by 'percli lint' before it can be deployed, preventing the application of malformed or malicious specifications.
Audit Metadata