groove-utilities-memory-install

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes mkdir -p using a path derived from a local configuration file. While a standard setup task, this pattern is a known attack surface if the configuration file is modified by an external actor to include shell metacharacters.- [PROMPT_INJECTION]: The skill exhibits an indirect injection surface by interpolating untrusted data from a file into a shell script.
  • Ingestion points: .groove/index.md (frontmatter memory: field).
  • Boundary markers: Absent; the value is used directly in the shell command template.
  • Capability inventory: Bash execution, file system read/write/glob/grep.
  • Sanitization: Absent; the skill does not perform validation or escaping of the path variable before shell execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 09:04 AM