litium-developer

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill has the capability to search and read content from the Litium community forum (forum.litium.com). This represents an indirect prompt injection surface where community-contributed content could potentially contain instructions for the agent.
  • Ingestion points: Forum search results and topic details are fetched via the Discourse API as described in references/forum-api.md.
  • Boundary markers: The skill does not explicitly define delimiters or specific 'ignore' instructions for forum content, treating it as reference material for 'community solutions'.
  • Capability inventory: The skill facilitates file system operations and execution of local development tools (dotnet, npm, yarn, npx) and platform APIs.
  • Sanitization: Content is ingested via standard JSON API responses.
  • [EXTERNAL_DOWNLOADS]: The skill references and uses external resources from Litium's official infrastructure.
  • Registries: It configures the @litiumab scope for npm and references Litium NuGet feeds at nuget.litium.com and packages.litium.com.
  • CLI Tools: It utilizes npx @litiumab/platform-extension-sdk for project scaffolding and management.
  • [COMMAND_EXECUTION]: The skill provides instructions for standard development workflows involving shell commands.
  • Standard Tools: Includes usage of dotnet, npm, yarn, npx, and curl for interacting with local development environments and platform APIs.
  • Developer Workflow: Facilitates administrative tasks such as database updates (dotnet litium-db), project creation (dotnet new), and client-side builds (yarn run prod).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 03:20 PM
Security Audit — agent-trust-hub — litium-developer