Security audit
Security checks for vulnerabilities and agentic risk
This PagerDuty skill needs review because its safety rules allow incident resolution without confirmation.
Before use, require confirmation for incident resolution and every other state-changing action, with an explicit allowlist for reads. Install the CLI through a reviewed, verified method and consider the access granted to OOMOL. The bundled instructions were fully reviewed, but the remote installer and live connector implementation were not verified.
The supplied code performs remote script execution rather than any visible PagerDuty operation. Nothing in the chunk establishes that this installer is a supporting PagerDuty dependency. The downloaded script's contents and resulting actions are not available for inspection.
Incident resolution is incorrectly covered by the instruction allowing untagged actions to run without confirmation. In an operational PagerDuty integration, unauthorized resolution can end active incident handling or suppress further escalation; this is the same underlying defect identified by SDI-4, not an additional independent vulnerability.
The setup command downloads a mutable remote script and immediately executes it without integrity verification or review; the PowerShell alternative has the same weakness. Although this is a legitimate dependency-installation context and runs only when the CLI is missing, compromise of the distribution endpoint could execute arbitrary code with the agent's local permissions.
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 resolve_incident action changes incident state but has no write or destructive tag, while the safety instructions explicitly classify untagged actions as reads that can run directly. This creates a confirmation bypass that could prematurely resolve an active incident and disrupt incident response.
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