memory-manager
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
lsandmvcommands to track and organize task folders within the project path. These operations are used to manage project metadata and directory structure. - [PROMPT_INJECTION]: The skill ingests text from project-level markdown files to update summary documents, representing an indirect prompt injection surface.
- Ingestion points: Reads content from
{project_path}/project_state.mdand task tracking files. - Boundary markers: Absent; the skill does not use specific delimiters for ingested content.
- Capability inventory: Includes file system read/write access and shell command execution.
- Sanitization: No validation or escaping is applied to the content before it is interpolated into project updates.
Audit Metadata