agent-log
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to search the
.agents/log/directory using therg(ripgrep) utility. While this is a standard operations tool, it relies on interpolating user-provided keywords into a shell command, which represents a potential attack surface for argument injection if the agent's tool execution environment does not perform adequate sanitization.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and follows instructions from the repository's.agents/log/directory, explicitly treating entries marked as 'done' as 'binding constraints'.\n - Ingestion points: The agent reads the
index.mdfile and individual markdown entry files within the.agents/log/folder.\n - Boundary markers: There are no explicit instructions to use delimiters or to disregard potential instructions embedded within the log content.\n
- Capability inventory: The agent possesses the capability to read and write files and execute shell commands.\n
- Sanitization: The skill lacks explicit sanitization or validation logic for the content retrieved from the logs before it is added to the agent's context.\n- [DATA_EXFILTRATION]: The skill manages internal repository logs and does not contain patterns for network exfiltration. It specifically includes a guideline advising the agent not to record sensitive data, such as secrets or credentials, which mitigates the risk of accidental data exposure.
Audit Metadata