hub-editor
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes hub configuration files (JSON) fetched from a remote service, which constitutes an ingestion of external data that could potentially contain malicious instructions.\n
- Ingestion points: The skill fetches envelope.json and extracts body.json via the dxs configuration get hub command (SKILL.md, references/hub-config-api.md).\n
- Boundary markers: The skill instructions lack explicit delimiters or specific "ignore embedded instructions" warnings for the data being processed.\n
- Capability inventory: The skill uses dxs configuration upsert hub, dxs configuration get hub, and dxs source explore to interact with platform configurations (SKILL.md, references/hub-config-api.md).\n
- Sanitization: Verification is performed via mandatory syntactic checks using python -m json.tool and manual review through diff -u between original and modified configuration bodies (SKILL.md).\n- [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell utilities including jq, diff, python, and the vendor-specific dxs CLI to manage and synchronize platform configurations. These operations are aligned with the skill's stated purpose of editing hub configurations.
Audit Metadata