Raon OS

PassAudited by ClawScan on May 10, 2026.

Overview

Raon OS appears purpose-aligned for startup plan analysis, but it uses local scripts, optional API keys, a local/webhook server, and stored conversation history that users should configure carefully.

Install only if you are comfortable with a local CLI/server tool processing startup documents. Configure only the API keys you intend to use, protect ~/.openclaw/.env, set the Kakao webhook secret before exposing the server, and review/clear stored history when working with confidential business plans.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Using the skill can run local Python/shell scripts against files you select, such as business-plan PDFs.

Why it was flagged

The package entry point executes the bundled raon.sh script with user-provided CLI arguments. This is command execution, but it is disclosed and central to the skill's CLI workflow.

Skill content
const result = spawnSync(sh, process.argv.slice(2), { stdio: "inherit", shell: false, });
Recommendation

Run it only from a trusted install source, keep the package updated, and avoid passing untrusted file paths unless you intend the skill to process them.

What this means

If configured, provider API keys may be used to send prompts or business-plan content to the selected LLM or managed API.

Why it was flagged

The skill expects optional LLM, Kakao, and managed-service API keys. This is normal for the stated integrations, but those credentials can authorize provider usage and spending.

Skill content
API keys are stored in ~/.openclaw/.env (user-managed, chmod 600 recommended).
Recommendation

Use least-privilege keys where possible, keep ~/.openclaw/.env readable only by you, and remove keys you do not intend this skill to use.

What this means

Business ideas, plan details, or user messages may be retained locally or in configured external storage.

Why it was flagged

The Kakao integration retains per-user conversation history, and the README also describes evaluation history and feedback storage. This persistence is disclosed but can involve sensitive startup information.

Skill content
세션 관리 — user.id 기반으로 대화 히스토리 유지 (최대 20턴)
Recommendation

Do not submit confidential material unless you are comfortable with the configured retention and provider settings; clear local history or disable external storage if needed.

What this means

If the server is exposed publicly, external callers may reach webhook endpoints and potentially trigger LLM/API usage.

Why it was flagged

The documentation supports exposing the local server through ngrok for Kakao webhooks and describes the HMAC secret as optional. This is a normal integration path, but public exposure without a secret can invite unwanted requests.

Skill content
KAKAO_CALLBACK_SECRET=your-secret  # 선택사항 ... ngrok http 8400
Recommendation

Set KAKAO_CALLBACK_SECRET, keep admin/API endpoints behind localhost or a firewall, and avoid exposing port 8400 broadly unless required.