Security audit
Security checks for vulnerabilities and agentic risk
Convex changes require approval, but setup runs unverified downloads and does not disclose the CLI’s automatic changes to agent skills.
Review and approve CLI installation separately. Verify the downloaded release and consider setting OO_SKILLS_SYNC_DISABLED=1 before running oo to prevent automatic skill synchronization, as documented in its [environment settings](https://github.com/oomol-lab/oo-cli/blob/main/docs/commands.md#environment-variables). Connect only the Convex permissions needed and retain mutation approvals. The installer and hosted connector code remain unverified.
The supplied command installs software from an Oomol endpoint and shows no Convex data operations or demonstrated connection to the declared purpose. Executing this unrelated remote installer is a materially different capability, not an established supporting implementation detail. The downloaded script's further behavior cannot be determined from the supplied chunk.
The only supplied behavior is authentication through oo; nothing identifies Convex as the target or implements the declared data operations. Authentication could be a supporting implementation detail if linked to Convex, but no such connection is present in this chunk.
Piping a remotely fetched script directly into bash executes mutable third-party code with the invoking user's permissions, without a separate integrity check or review step. Compromise of the distribution endpoint or installer could expose local secrets or alter the environment; HTTPS and a legitimate setup purpose do not eliminate this supply-chain 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 missing-CLI fallback directs installation through immediate execution of downloaded shell or PowerShell code without explaining the local execution or obtaining explicit installation approval. Authorization to operate Convex does not necessarily authorize modifying the local environment; the failure-triggered setup limits exposure but does not supply that consent.
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