far-general
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references multiple external resources for SDK quickstarts and expanded documentation.
- Instruction files point to official vendor repositories on GitHub, such as
github.com/Banuba/quickstart-webandgithub.com/Banuba/banuba-sdk-android-samples. - The agent is instructed to fetch
llms-full.txtfrom the official domaindocs.banuba.comif local documentation is missing. - Bundled documentation files load the Mermaid diagramming library from the well-known CDN
cdn.jsdelivr.net. - [PROMPT_INJECTION]: The skill instructions define logic for the agent to ingest and analyze local project files, which creates a surface for indirect prompt injection (Category 8).
- Ingestion points: The agent reads project configuration files like
package.json,build.gradle,pubspec.yaml, and source files (*.swift,*.kt,*.cpp) to detect the development platform and integration context. - Boundary markers: The instructions do not implement specific delimiters or safety instructions to prevent the agent from following malicious commands that might be embedded in these third-party project files.
- Capability inventory: The skill possesses significant capabilities, including project scaffolding and generating implementation code for Web, Android, iOS, and other platforms.
- Sanitization: No explicit sanitization or filtering is prescribed for the data ingested from the local environment before it is used in the agent's reasoning process.
Audit Metadata