Security audit
Security checks for vulnerabilities and agentic risk
The searches fit the skill’s purpose, but a routine request can trigger an unverified software installer without asking you first.
Review the setup behavior before installing. Preinstall a reviewed, version-pinned CLI and require approval before any installer fallback runs. Check whether the CLI will install or update other agent skills, and understand that OOMOL mediates access to your connected account. The Review verdict concerns insufficient control over installation; the inspected artifact does not establish malicious intent.
The supplied command performs remote installer execution, not a The Swarm search or read operation. No connection is shown between this installer and the declared service, so it cannot be established as a supporting implementation detail. The installer contents are not supplied, so its further effects cannot be determined.
Authentication can be a supporting implementation detail, but the supplied chunk consists entirely of a login command and provides no evidence that it authenticates to The Swarm or supports the declared data-access functionality. Its demonstrated primary behavior therefore differs from the declared purpose.
The command pipes a mutable remote script directly into Bash without a pinned version or independent integrity verification. If the distribution endpoint or release infrastructure is compromised, attacker-controlled code executes with the agent's local privileges; HTTPS alone does not protect against a compromised publisher. The PowerShell installation example has the same underlying 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
When the CLI is missing, the skill instructs the agent to execute a downloaded installer without first obtaining approval for local software installation. A read-only service request can therefore lead to code execution and system changes under the agent's local permissions; the missing-command condition limits when this happens but does not provide informed authorization.
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