lov-obsidian-ensure-dev-sync

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: Analysis of the skill instructions and associated scripts reveals a focus on secure environment management. The utility scripts use safe defaults for file permissions and validate input segments before processing.
  • [INDIRECT_PROMPT_INJECTION]: The skill maintains an attack surface common to development tools by processing external project configurations.
  • Ingestion points: Reads and interprets Obsidian plugin manifests (manifest.json), build configuration files (e.g., esbuild.config.js), and development sync hooks within target project directories.
  • Boundary markers: The skill explicitly separates discovery, preview, and execution phases, ensuring modifications only occur within user-defined scopes.
  • Capability inventory: The skill performs file-system operations including directory synchronization, file copying, and local build script invocation through host capabilities.
  • Sanitization: Instructions mandate the use of parameterized paths and explicitly prohibit shell string concatenation in execution hooks to mitigate injection risks.
  • [DATA_EXFILTRATION]: The scripts/profile_store.py script manages local configuration records and implements a security filter (SENSITIVE_PARTS) to block the persistent storage of credentials, tokens, and secrets in the shared profile.
  • [COMMAND_EXECUTION]: The workflow documentation specifically cautions against insecure shell usage (execSync) and requires result verification after each synchronization step, reinforcing a safe execution boundary.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 09:49 AM
Security Audit — agent-trust-hub — lov-obsidian-ensure-dev-sync