skills/mukul975/anthropic-cybersecurity-skills/analyzing-security-logs-with-splunk/Gen Agent Trust Hub
analyzing-security-logs-with-splunk
Warn
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Potential SPL injection vulnerability in
scripts/agent.py. The script constructs search queries using f-strings and direct string concatenation with values provided via command-line arguments. Specifically, thebuild_incident_timelinefunction joins user-provided host and user names with" OR "and wraps them in double quotes. Since these inputs are not sanitized or escaped, a user could provide a crafted host or username containing a closing quote and SPL command pipe (e.g.,WKSTN-042" | delete | search host=") to execute unauthorized actions within the Splunk environment. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted security logs that could contain malicious instructions designed to mislead an agent or analyst.
- Ingestion points: The agent script
scripts/agent.pyingests data from multiple Splunk indexes includingwindows,sysmon,proxy, andfirewall. - Boundary markers: The SPL queries in
SKILL.mdandscripts/agent.pyuseevalto create summaries but do not implement boundary delimiters to separate the log content from the analysis context. - Capability inventory: The skill has the ability to query Splunk APIs and generate reports based on the retrieved data.
- Sanitization: There is no evidence of sanitization, filtering, or escaping of log content before it is interpolated into the final report structure.
Audit Metadata