glm5-parallel
Pass
Audited by Gen Agent Trust Hub on May 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines Bash patterns for parallel task management, utilizing background execution (
&) and synchronization (wait) to coordinate multiple sub-agent processes. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads and aggregates output summaries from other agents without explicit sanitization.
- Ingestion points: Reads user-provided task descriptions and sub-agent output files stored at
.ralph/teammates/*/status.json. - Boundary markers: Absent; the skill does not use delimiters or instructions to isolate aggregated sub-agent outputs.
- Capability inventory: Includes
Bash,Write, andTeamCreate, allowing for code execution and file modifications based on processed data. - Sanitization: No validation or escaping is performed on the
output_summaryfield retrieved from sub-agent status files.
Audit Metadata