Security audit
Security checks for vulnerabilities and agentic risk
This Cohere helper needs review because setup executes changeable remote code and its dependency can install persistent skills into other agent applications.
Review the CLI installation before allowing automatic setup. Use a verified release, check which additional agent skills it installs, and consider disabling startup skill synchronization with OO_SKILLS_SYNC_DISABLED=1 before connector calls. Send only content you authorize OOMOL and Cohere to process. The bundled skill was fully inspected, but remote installer contents and live connector implementations were not verified.
The supplied code installs software from an Oomol-hosted endpoint rather than performing any visible Cohere operation. Dependency installation can be a supporting implementation detail, but this chunk establishes no relationship between the installer and the declared Cohere functionality. The remote script's contents and further effects cannot be determined from the supplied code.
The setup command pipes a mutable remote script directly into bash without a pinned version, integrity verification, or review; the PowerShell alternative has the same execution pattern. Although this is a legitimate dependency-installation step triggered only when the CLI is missing, compromise of the distribution endpoint or release pipeline could execute arbitrary commands with the agent's local privileges and expose accessible files or credentials.
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 execute Cohere actions through OOMOL, including chat and document reranking, which transmit supplied prompts, queries, or documents to external services. Although the integration is identified, there is no warning about this disclosure, and the safety guidance treats untagged actions as reads that are safe to run directly.
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