Security audit
Security checks for vulnerabilities and agentic risk
The Meituan search workflow fits its purpose, but fallback setup executes unverified downloaded code without a clear approval step.
Review the fallback setup before installing. Install a pinned, verified CLI release yourself and require approval for subsequent installation changes. Check the CLI's automatic skill synchronization settings. Meituan requests use your OOMOL-connected account and may consume credit. The bundled artifact shows no malicious behavior, but the downloaded installers remain uninspected.
The supplied code only invokes a login command and shows no connection to Meituan or implementation of the declared search and reading functions. Authentication could support such functionality, but the supplied chunk provides no evidence that this login is a supporting step for Meituan access.
The skill directs the agent to install software by executing remotely fetched scripts when the CLI is missing, expanding a data-reading task into local code execution. Conditional setup makes this less suspicious than unconditional execution, but it does not require explicit installation approval or verify the downloaded code; this overlaps with the installer findings below.
The curl-to-bash command executes whatever script the remote endpoint serves without pinning or integrity verification. HTTPS protects transport but not against a compromised publishing endpoint or malicious script update, which could yield arbitrary code execution under the agent's local account.
oo: command not found — install the oo CLI (other platforms: https://cli.oomol.com/install-guide.md):
curl -fsSL https://cli.oomol.com/install.sh | bash # macOS / Linux
Both setup examples execute downloaded scripts directly without an inspection, integrity-verification, or explicit approval step. Such scripts run with the invoking user's privileges and could modify files or access credentials; the missing-CLI condition limits exposure but does not constrain their effects.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
- **`OOMOL_INSUFFICIENT_CREDIT`** — OOMOL billing stop. Recharge at `https://console.oomol.com/billing/token-recharge` before retrying.
- **HTTP 402 without `OOMOL_INSUFFICIENT_CREDIT`** — the third-party account itself has no active plan or credit. Report the provider error message as is; do not ask the user to recharge OOMOL.
## Resources