Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Ai Director
v0.1.1AI short drama generation - account management, script writing, video production. Integrated X2C billing for commercial deployment.
⭐ 0· 183·1 current·1 all-time
byParker@patches429
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The declared purpose (AI short drama generation + X2C billing) aligns with requiring node and an X2C API key. However the documentation frequently references other integrations (Feishu document upload, Gemini QC, various helper scripts like giggle-api.js/auto-generate.js and Python drama_manager.py) that are not present in the file manifest. The skill both advertises per-user credential files (credentials/{USER_ID}.json) and also requires a global X2C_API_KEY environment variable — the need for a global service key is not justified by the stated multi-user binding flow.
Instruction Scope
Runtime instructions tell the agent to read/write local credential files (credentials/{USER_ID}.json) and create projects/versions under a projects/ folder (writing to disk). They also direct many network interactions with external APIs (X2C/giggle.pro). The SKILL.md references env vars (USER_ID, TELEGRAM_USER_ID) and other services (GEMINI_API_KEY, Feishu) that are not declared as required in the registry metadata; scripts may therefore access additional secrets or endpoints not listed. Documentation commands reference several script names that are missing from the manifest, so following the SKILL.md blindly could result in failed or unexpected operations.
Install Mechanism
No install spec (instruction-only) is supplied; scripts are plain JS files to be executed with node. This is lower risk than downloading and extracting arbitrary binaries. There is no installer that would fetch code from unknown URLs in the manifest provided.
Credentials
The registry requires a single env var X2C_API_KEY which is appropriate for the advertised X2C integration. But the skill also mentions optional GEMINI_API_KEY (for quality scoring) and Feishu upload in the docs without declaring them as required. The coexistence of a required global X2C_API_KEY plus per-user credential files is inconsistent: if per-user API keys are the norm, a global X2C_API_KEY may be unnecessary and could be misused. The scripts will accept/persist keys to credentials/{USER_ID}.json — storing user API keys locally in plaintext increases risk of exposure to other local processes or skills.
Persistence & Privilege
The skill writes project data and user credential files to disk (projects/, credentials/) and updates local state; this persistence is expected for a production workflow. always:false and normal autonomous invocation are in place. Still, persisting per-user API keys and auto-saving project versions means the skill will retain credentials and created assets on disk — users should be aware and control file permissions and retention.
What to consider before installing
Things to check before installing or enabling this skill:
- Confirm why a global X2C_API_KEY is required when the skill supports per-user API keys (credentials/{USER_ID}.json). Prefer using per-user keys or a dedicated low-privilege test key rather than your main account key.
- The docs mention uploading to Feishu and using GEMINI_API_KEY for QC but those credentials are not declared as required; inspect the actual JS files (ad-account-manager.js, ad-producer.js, quality-evaluator.js) to see whether they call external endpoints that need additional secrets.
- The repository docs reference many scripts that are not present (Python drama_manager.py, giggle-api.js, auto-generate.js). That mismatch could mean the docs are stale or the skill is incomplete — don't assume all advertised behavior exists.
- Per-user credentials are stored as plaintext JSON under credentials/{USER_ID}.json. If you must use it, store them in a sandbox, restrict filesystem permissions, and avoid placing org-level or long-lived keys there.
- Review network endpoints in the scripts (e.g., giggle.pro / api.giggle.pro / X2C base URL) to confirm they are expected and trustworthy. Consider running the skill in an isolated environment (VM/container) first and testing with a throwaway X2C account.
- If you rely on the Gemini quality-evaluator, verify whether the skill will send video thumbnails or full video URLs to an external QC service and whether that service requires a separate API key.
- If anything is unclear, ask the skill author for a manifest of actual runnable scripts and a mapping of which env vars each script reads (and where credentials are persisted).scripts/ad-account-manager.js:65
Shell command execution detected (child_process).
scripts/auto-iterate.js:108
Shell command execution detected (child_process).
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.Like a lobster shell, security has layers — review code before you run it.
latestvk978cxx3399jqn0zaz4xxcr6gn82vk0h
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🎬 Clawdis
Binsnode
EnvX2C_API_KEY
Primary envX2C_API_KEY