household-waste-info
Fail
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill requires the execution of an external CLI tool using
npx -y @nomadamas/k-skill@0. This command downloads and runs code from the npm registry at runtime without pre-verification. - [EXTERNAL_DOWNLOADS]: The skill relies on external resources hosted at
github.com/NomaDamas/k-skillandnomadamas.orgfor its core instructions and updates. - [COMMAND_EXECUTION]: The instructions mandate running shell commands (
npx,curl) to function, fetch instructions, and update the skill. This bypasses static instruction definitions. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from a public API via a proxy (
k-skill-proxy.nomadamas.org), which could be manipulated by an attacker to include instructions for the agent. - Ingestion points: API responses from the proxy server fetched via
curlininstruction.md. - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded commands in the API response.
- Capability inventory: The agent has access to
curlandnpxcommands. - Sanitization: No specific sanitization or escaping of the API response content is defined beyond a request to summarize the information.
Recommendations
- AI detected serious security threats
Audit Metadata