Security audit
Security checks for vulnerabilities and agentic risk
This skill needs review because setup directly executes unverified remote code and its safety rules treat potentially paid AI generation as a read-only action.
Review and verify the CLI installation before allowing automatic setup, and check its skill-synchronization settings because it can affect other agent installations. Authorize chat generation deliberately, with suitable spending limits, and submit only content you intend the OOMOL/Eden AI integration to process. The complete bundled artifact was inspected; remote installer and hosted-service behavior remain unverified. No malicious behavior was established.
The supplied command contains no Eden AI search or read operation. Instead, it fetches and executes an installer from a different service. This introduces an undeclared installation and remote-code-execution capability, with no supplied evidence that it is a supporting implementation detail for Eden AI access. The downloaded script's further behavior cannot be determined from this chunk.
The supplied chunk only initiates CLI login; it contains no Eden AI search, read, or integration behavior. Authentication could be a supporting implementation detail if its relationship to Eden AI were established, but that relationship is not demonstrated here. The command alone does not establish credential theft or other malicious behavior.
The fallback installer pipes a mutable remote script directly into Bash without version pinning, integrity verification, or review; the PowerShell alternative has the same execution pattern. Compromise of the hosting service or release pipeline could execute arbitrary code with the agent's local privileges, exposing accessible files and credentials; conditional execution reduces frequency but not the potential impact.
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 untagged create_chat_completion action inherits the blanket instruction that untagged actions are reads and safe to run directly, although it invokes inference and can consume paid credits. This classification can lead an agent to perform billable generation under a request framed as searching or reading, bypassing the skill's safeguards for actions with effects.
The workflow instructs the agent to execute connector actions with a JSON payload, including the listed chat-completion action. Although this sends prompt or conversation content to Eden AI, the skill provides no privacy warning and instead classifies 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