skillshare-devcontainer
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONMETADATA_POISONINGDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands and scripts (e.g., '.devcontainer/ensure-skillshare-linux-binary.sh') inside a Docker container via 'docker exec'. This includes complex multi-command sequences and administrative tasks like starting/stopping the container.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection. 1. Ingestion points: Repository source code and files in '/workspace', command outputs from 'docker exec'. 2. Boundary markers: Absent; no specific delimiters are used to wrap external content. 3. Capability inventory: Arbitrary shell execution via 'docker exec', file system access within the container/bind-mounted workspace, and network access for the Web UI dashboard. 4. Sanitization: Absent; no mention of filtering or escaping external content before interpolation into shell commands.
- [METADATA_POISONING]: There is a discrepancy between the declared author 'Runkids' and the expected vendor context 'jetbrains'. The source URL points to a GitHub repository ('runkids/skillshare') that is not associated with recognized vendor infrastructure.
- [DYNAMIC_EXECUTION]: The skill dynamically generates and executes shell scripts and Go tests within the container environment based on high-level task descriptions provided to the agent.
Audit Metadata