ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Snap Illustrator

v1.0.5

Instantly analyze markdown articles, generate prompts, and seamlessly insert AI illustrations with zero configuration required. Supports Pollinations API (ze...

⭐ 0· 89·0 current·0 all-time
by@wysaid
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
!
Purpose & Capability
The SKILL.md repeatedly promotes a zero‑config flow using the unauthenticated Pollinations endpoint, yet the registry metadata marks HF_TOKEN as a required environment variable. Requesting HF_TOKEN as 'required' is disproportionate given the primary flow does not need it; the included script only uses HF_TOKEN as a fallback or from a config file.
!
Instruction Scope
Instructions correctly restrict generation to the provided markdown and running the bundled Node script, but they also tell the agent to modify the skill's SKILL.md frontmatter to persist an HTTP proxy and to automatically reuse that proxy in subsequent runs. That instructs the agent to edit the skill bundle on disk and to silently apply stored settings later, which broadens scope beyond simple image generation.
✓
Install Mechanism
There is no install spec and only a small Node script is included. The skill requires the 'node' binary but does not download code from external URLs or perform any extract/install steps — this is low risk from an installation perspective.
!
Credentials
The skill's registry metadata lists HF_TOKEN as required, but the SKILL.md and script treat HF_TOKEN as optional/fallback. The script will also read a config file under the user's HOME (~/.config/snap-illustrator/config.json). Asking for HF_TOKEN up front (or marking it required) is disproportionate to the stated zero‑config behavior and increases credential exposure risk if users provide tokens without understanding where they will be used or stored.
!
Persistence & Privilege
The skill does not set always:true, but the instructions explicitly ask the agent to persist a user‑provided HTTP proxy by editing SKILL.md frontmatter so the proxy is reused automatically in future runs. Persisting settings by modifying the skill files (rather than asking the user each time or using a clearly scoped config store) is a privilege escalation in practice and should be treated cautiously.
What to consider before installing
What to consider before installing: - The core code (scripts/generate.mjs) is small and implements the advertised behavior: it tries Pollinations (unauthenticated) and falls back to HuggingFace if an HF token is available. - However, the registry metadata declaring HF_TOKEN as required is inconsistent with the SKILL.md's zero‑config promise — don't assume you must supply a token to use the skill. - The runtime instructions tell the agent to edit the skill's SKILL.md to store an HTTP proxy and to reuse it automatically later. Be cautious: this persists configuration into the skill bundle and enables future automatic behavior without explicit per‑run consent. - If asked for a HuggingFace token, only provide it if you trust the skill and understand that the script will use it to call HuggingFace APIs; avoid pasting secrets into chat. Prefer to run the Node script yourself locally so you control where tokens are stored. - If you install: (1) review/approve any changes before the skill edits files, (2) verify that the skill will not store your token unencrypted in global config unless you explicitly ask, and (3) consider running the included script manually in a sandboxed environment to verify outputs first. - If you want to proceed, ask the author to remove HF_TOKEN from the 'required' metadata (make it optional/fallback) and to stop recommending editing SKILL.md for proxy storage; store per‑user proxy settings in a clearly scoped config with explicit user consent instead.
✗
scripts/generate.mjs:79
Environment variable access combined with network send.
!
scripts/generate.mjs:83
File read combined with network send (possible exfiltration).
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.

latestvk974chbtb0q5s5r1gj5g1s43yn83cp5y

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

Binsnode
EnvHF_TOKEN

Comments