lx-ppt

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the lx command-line interface to perform operations such as generating, modifying, and deleting PPT pages. These are standard functional calls to the vendor's provided tooling.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data (user-provided context, planning descriptions, and external reports) which is then passed as arguments to AI-driven commands (lx ppt generate-ppt). This is an inherent risk of the functionality but no malicious exploitation patterns were found in the static code.
  • Ingestion points: Data enters through the --context, --planning, and --deep-research-report-url parameters in SKILL.md and references/ppt.md.
  • Boundary markers: None explicitly defined to separate user data from instructions in the command arguments.
  • Capability inventory: Subprocess execution of lx CLI commands across all operational scripts.
  • Sanitization: No explicit sanitization or escaping of input data is defined within the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 09:42 PM
Security Audit — agent-trust-hub — lx-ppt