Security audit
Security checks for vulnerabilities and agentic risk
The Yunbiao workflow includes clear approval safeguards, but its setup fallback runs remote installer code without a separate approval step or integrity check.
Review and approve CLI installation separately, preferably using a pinned, verified release, before allowing the fallback to run. Connect only the Yunbiao account and permissions needed, and retain the required approvals for business-data and account changes. The entire bundled skill was inspected, but the external installer and connector implementation remain unverified; no malicious behavior was established.
The supplied chunk runs an Oomol-hosted installer, while the description declares Yunbiao data management. Nothing in the supplied code establishes that this installation supports Yunbiao, so it cannot be treated as a demonstrated supporting implementation detail. The remote script's contents and downstream effects are not available for assessment.
Piping a remotely fetched script directly into Bash executes whatever the installation endpoint returns with the invoking user's privileges, without a pinned version or independently verified integrity. A compromised vendor endpoint or distribution pipeline could therefore execute arbitrary local code; HTTPS does not protect against compromise of the publisher. The PowerShell example has the same risk, although installation is only a fallback and the domain belongs to the expected CLI provider.
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
The missing-CLI fallback directs the agent to execute an installer without explicitly obtaining permission for local code execution or explaining its system changes. Permission to perform Yunbiao operations does not necessarily authorize installing software, and the write/destructive confirmation rules only address Yunbiao state. The conditional fallback reduces exposure but does not resolve this consent gap.
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