agent-ready-cli
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using
npxandnpmto run theagent-ready-scannerutility. - [EXTERNAL_DOWNLOADS]: The skill fetches the
agent-ready-scannerpackage from the npm registry for execution. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites during scans, creating an attack surface where malicious instructions could be embedded in site content like page titles, headings, or
llms.txtfiles. - Ingestion points: Website content retrieved and displayed by the
agent-ready scancommand (SKILL.md). - Boundary markers: Present; the skill includes a "Security & trust" section that explicitly instructs the agent to treat scan results as inert data and ignore embedded instructions.
- Capability inventory: Shell command execution (
npx,npm,grep,printenv) and file system access (reading.envfiles). - Sanitization: Absent; the skill relies on instructional guardrails rather than automated sanitization of the scanned content.
Audit Metadata