cloudflare-safe
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Node.js scripts (
cf-read.mjs,cf-apply.mjs,setup-tokens.mjs) to manage Cloudflare resources and configure local credentials. These operations are restricted to the skill's directory and intended functionality. - [DATA_EXFILTRATION]: The skill handles sensitive API tokens stored in
~/.config/cloudflare/tokens.env. It employs multiple safeguards to prevent exfiltration, including redacting tokens from all tool outputs and strictly validating that API calls are only sent to the official Cloudflare domain (api.cloudflare.com). - [INDIRECT_PROMPT_INJECTION]: The skill reads configuration data from Cloudflare APIs. Although this data is externally sourced, the risk of indirect prompt injection is mitigated by the skill's requirement for human oversight on all changes and the agent's internal safety protocols.
- [SAFE]: The skill's design prioritizes security by keeping credentials out of the agent's context and directing all network activity to a well-known, trusted service provider.
Audit Metadata