Skill flagged — suspicious patterns detected

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

TrendProof

v1.2.2

Query TrendProof (trendproof.dev) for keyword trend velocity scores. Returns velocity score, trend direction (rising/stable/falling), monthly search volume,...

0· 117·0 current·0 all-time
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's name/description (query TrendProof) matches the included CLI script which performs HTTP calls to trendproof.dev. However the registry metadata declared no required env vars or config paths while the code actually reads TRENDPROOF_API_KEY and TRENDPROOF_BASE_URL and reads/writes a config file at ~/.config/clawdbot/trendproof.json. That mismatch between declared requirements and actual behavior is an incoherence the user should notice.
!
Instruction Scope
SKILL.md instructs the agent to check for an API key and — if missing — to ask the user to "Copy your key ... and send it to me." That phrasing encourages users to paste a secret directly into the chat session. The script also documents running a configure command which writes the key to disk; both behaviors are within the stated purpose (calling the API) but create clear opportunities for sensitive data exposure if the user follows the 'send it to me' path. Otherwise the instructions stick to API calls and expected CLI usage.
Install Mechanism
No install spec (instruction-only + a pure-Python script included). The script uses only the Python standard library and does not download external code, so installation risk is low.
!
Credentials
Providing an API key (TRENDPROOF_API_KEY) is proportionate to the skill's function, but the metadata advertised no required env vars while the code relies on them. The script also accepts TRENDPROOF_BASE_URL as an override — useful for testing but potentially dangerous because a malicious or misconfigured base URL would cause the skill to send the API key and requests to an arbitrary endpoint. The script persists the key to ~/.config/clawdbot/trendproof.json, which is expected behavior but should be disclosed up-front in the metadata.
Persistence & Privilege
The skill is not always: true and does not request elevated privileges. It does persist user API keys to a local config file and supports an override env var for base URL. Autonomous invocation (default) is allowed — normal for skills — so consider that the agent could use a stored key without further prompts.
What to consider before installing
This skill appears to implement the advertised TrendProof API calls, but there are a few things to watch before installing or using it: - Do not paste your API key into chat. SKILL.md's wording encourages users to "send it to me"; instead run the provided configure command locally (python3 skills/trendproof/scripts/trendproof.py configure --api-key TRND_xxxxx) or set TRENDPROOF_API_KEY in your shell. Pasting secrets into a conversation can expose them to logs or the agent's memory. - Expect the script to write the key to ~/.config/clawdbot/trendproof.json. If you prefer not to store the key on disk, use an environment variable for the session and remove it when done. - The script allows overriding the API base URL via TRENDPROOF_BASE_URL. Leave this unset unless you trust the endpoint; a malicious override would cause your API key and requests to be sent to an arbitrary server. - Metadata omitted the config/env requirements; if you want to be cautious, review the included scripts/trendproof.py yourself (or have a trusted developer do so) before supplying credentials and confirm the owner/source of the skill (source is listed as unknown). - If you still want to use it: prefer setting TRENDPROOF_API_KEY in your environment or running the configure command locally rather than pasting the key into agent chat, and review the saved config file to verify only the expected key is stored.

Like a lobster shell, security has layers — review code before you run it.

latestvk97e0he90w1wk5v1s8j70bcz8h833432

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments