lokalise-upgrade-migration

Warn

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: MEDIUMPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The skill uses sudo in its implementation examples to move a downloaded binary into system executable directories (e.g., /usr/local/bin/).
  • [COMMAND_EXECUTION]: Multiple Bash scripts are included to perform version audits using npm and node, locate code patterns using grep, and manage git branches.
  • [EXTERNAL_DOWNLOADS]: Fetches the latest CLI release and version metadata from the official Lokalise GitHub repository.
  • [DYNAMIC_CONTEXT_INJECTION]: The skill uses !command syntax in the SKILL.md header to execute shell commands at load time, such as checking for installed SDK versions and Node.js availability.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests local file data to inform the migration process, creating a surface for indirect instructions.
  • Ingestion points: Reads contents from package.json and source files via shell commands.
  • Boundary markers: The dynamic context injections do not use explicit delimiters to separate system output from instructions.
  • Capability inventory: The skill has permission to use Bash (npm, npx, node, git, grep), Write, and Edit tools.
  • Sanitization: No sanitization or validation is applied to the ingested file content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 9, 2026, 03:47 AM
Security Audit — agent-trust-hub — lokalise-upgrade-migration