concurrency-exploitation
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents numerous commands for conducting race condition attacks and system analysis, including process priority manipulation and CPU stressing.
- Evidence:
SKILL.mdandpayloads.mdlist tools likestress-ng,taskset, andstracefor manipulating system behavior. - [DATA_EXPOSURE]: Documentation explicitly references sensitive system paths as targets for exploitation examples.
- Evidence:
SKILL.mdandtest-cases.mduse/etc/shadowas a target example for symlink-based TOCTOU attacks. - [PRIVILEGE_ESCALATION]: The instructions include techniques for acquiring elevated privileges through process priority adjustment.
- Evidence:
guides/race-window-amplification.mdsuggests usingsudo nice -n -20to increase attacker process priority. - [INDIRECT_PROMPT_INJECTION]: The skill is configured to ingest external data from the web and system logs, creating a potential attack surface if malicious instructions are embedded in those sources.
- Ingestion points:
SKILL.mdallowsWebSearchandWebFetchtools;test-cases.mdandguides/race-window-amplification.mdinstruct the agent to monitoraudit logsanddmesg. - Boundary markers: Absent.
- Capability inventory: The skill has
BashandWritepermissions (SKILL.md). - Sanitization: Absent.
- [DYNAMIC_EXECUTION]: The skill guides the user to compile and execute code at runtime to detect or trigger race conditions.
- Evidence:
SKILL.mdandguides/threadsanitizer-guide.mdprovide patterns for compiling source code withgcc -fsanitize=threadand executing the resulting binary.
Audit Metadata