Caveat verdict
mmxagent-skill-wechat
Reads local files AND makes external network calls
Personal WeChat connection skill using official Tencent WeChat plugin; requires explicit user confirmation before writing credentials to disk, and the skill clearly describes what packages are installed and from where.
Automated static analysis — 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 — 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 — the capabilities for data exfiltration co-occur (data-flow not verified)
LLM02 · LLM06 · ASI03
Accesses credentials AND makes external network calls — potential credential theft
LLM02 · ASI03
Accesses credentials AND writes files — may persist stolen credentials locally
LLM02 · LLM06 · ASI03
Both reads and writes files — verify scope is limited to intended directories
LLM06 · ASI02
Permission integrity
network_out
file_read+write
credential_access
package_install
Findings (1)
Possible hardcoded credential
SKILL.md · code · token: '<bot_token>
Permissions & capabilities
No declared permissions — minimal attack surface.
package_installnetwork_outfile_readfile_writecredential_access Thanks — recorded.