flux-topology
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from the local file system (specifically YAML files within the
flux/directory) to construct a topology graph. This creates a surface for indirect prompt injection, where malicious instructions could be embedded within the GitOps configuration to influence the agent's behavior. - Ingestion points: Reads all YAML files located under the
flux/directory in the working directory. - Boundary markers: The instructions do not define explicit delimiters or instructions to treat the file content strictly as data, increasing the risk that the agent may follow instructions found within the repository.
- Capability inventory: Includes tools for mapping topology (
fluxtop_map), inspecting specific applications (fluxtop_app_card), tracing dependency chains (fluxtop_trace), and performing regex searches across all applications (fluxtop_find_refs). - Sanitization: No evidence of input sanitization, schema validation, or escaping of the processed external content is provided in the skill instructions.
Audit Metadata