Security audit
Security checks for vulnerabilities and agentic risk
The Jina AI workflow fits its purpose, but installation and automatic changes to other agent skills need review.
Review and verify the CLI before allowing installation. Check whether it automatically installs other agent skills, and disable synchronization if unwanted. Only submit content you intend to share with OOMOL and Jina AI. The package contains only instructions; the external installer and exact runtime release remain unverified.
The supplied command installs software from an Oomol-controlled endpoint rather than performing Jina AI operations. No connection to Jina AI or necessity as a supporting implementation detail is established in the supplied code. The downloaded script's further behavior cannot be determined from this chunk.
The supplied code performs only an oo CLI login and contains no Jina AI interaction or connection between that login and Jina AI. Authentication could be a supporting implementation detail if such a connection were established, but the supplied chunk instead presents a materially different standalone behavior.
The setup instructions pipe a mutable remote script directly into Bash without inspecting it or verifying a pinned version, checksum, or signature; the PowerShell alternative has the same execution pattern. If the distribution endpoint or release pipeline is compromised, an agent following this fallback could execute attacker-controlled code with its local permissions, exposing accessible credentials and files; the missing-command condition reduces exposure but does not remove the risk.
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 workflow instructs the agent to submit action payloads through OOMOL to Jina AI, including text, documents, and queries used for embedding or reranking. Although the services are named, the skill does not warn that supplied content leaves the user's environment, and it labels untagged actions safe to run directly despite their potential privacy impact.
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