honey-ccr
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted tool output (such as logs or scan results) and inserts retrieval instructions into the agent's context.\n
- Ingestion points: Processes output from arbitrary shell tools piped into the compression utility (referenced in SKILL.md).\n
- Boundary markers: No explicit markers or "ignore instructions" wrappers are mentioned to isolate the sampled tool data from the agent's instructions.\n
- Capability inventory: Executes subprocess commands via
eson retrieveto restore cached data (referenced in SKILL.md).\n - Sanitization: No sanitization of the tool output is described before it is sampled and displayed to the agent.\n- [COMMAND_EXECUTION]: The skill's primary workflow involves executing the
esonCLI tool for data compression and retrieval operations.
Audit Metadata