Caveat verdict
wechat-publish
Reads local files AND makes external network calls
WeChat article publishing workflow using official WeChat material and draft APIs; credential_access is for the user's own WeChat AppID/AppSecret and all network calls go to documented api.weixin.qq.com endpoints.
Automated static analysis β not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
What it does
These are capability combinations: each listed behavior occurs in the skill, but Caveat detects co-occurrence β it does not verify that one flows into another. Read the code to confirm a live chain.
Reads local files AND makes external network calls β the capabilities for data exfiltration co-occur (data-flow not verified)
LLM02 Β· LLM06 Β· ASI03
Accesses credentials AND makes external network calls β potential credential theft
LLM02 Β· ASI03
Permission integrity
network_out
file_read
credential_access
Findings (2)
Pipe to python β executes piped content as Python code
SKILL.md Β· code Β· | python3
Possible hardcoded credential
SKILL.md Β· code Β· TOKEN="θ·εε°ηtoken
Permissions & capabilities
No declared permissions β minimal attack surface.
network_outfile_readcredential_access Thanks β recorded.