unplugin-skilld
Warn
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: MEDIUMREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill recommends using
npx -y skilld search. The-yflag bypasses the interactive confirmation prompt for package installation, leading to the immediate download and execution of theskilldpackage from the NPM registry. - [EXTERNAL_DOWNLOADS]: Recommends downloading and running the
skilldpackage from the public NPM registry. This package is not a standard industry utility nor is it part of the verified namespace of the skill's author. - [COMMAND_EXECUTION]: The skill provides bash commands for the agent to execute in the local environment to search documentation and issue indexes.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from local project files, which creates a potential surface for indirect prompt injection if those files are attacker-controlled.
- Ingestion points: Files located in the
.skilld/directory, includingpackage.json,README.md, and indexed documentation/issues/releases. - Boundary markers: None present in the instructions to prevent the agent from following instructions potentially embedded within these indexed files.
- Capability inventory: The skill possesses command execution capabilities (
npx) and the ability to read and process local file content. - Sanitization: No evidence of sanitization or validation of the content retrieved from the
.skilld/directories before it is processed by the agent.
Audit Metadata