skills/wenerme/ai/opencode-docs/Gen Agent Trust Hub

opencode-docs

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a collection of documentation files assisting the agent in answering questions about OpenCode.
  • [PROMPT_INJECTION]: Some documentation files (e.g., references/ai-tools/cursor.mdx, references/ai-tools/windsurf.mdx) contain example system prompts and rules for other AI tools. These represent a potential surface for indirect prompt injection if the agent retrieves and adopts them as its own operating constraints. 1. Ingestion points: references/ directory via grep as instructed in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: The skill documentation describes full access to shell commands and file operations. 4. Sanitization: Absent.
  • [EXTERNAL_DOWNLOADS]: The documentation correctly references official installation scripts from the opencode.ai domain and standard package registries (NPM, Brew, etc.).
  • [COMMAND_EXECUTION]: The documentation includes standard examples of CLI usage for the tool, such as configuration updates and plugin management.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 08:05 PM
Security Audit — agent-trust-hub — opencode-docs