Wechat Db Decrypt
AdvisoryAudited by Static analysis on May 3, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Using this on a machine can expose the logged-in WeChat account's local message databases, including private and group chats.
The skill relies on an active WeChat session and extraction of an in-memory database key, while registry metadata declares no credential or config requirement and the artifacts do not clearly bound which account/data may be accessed.
前提：微信已正常登录 ... 它会自动找到微信进程、提取密钥、解密数据库
Use only on your own account and device, require an explicit user-selected wxid/database path, document exactly what key material is accessed, and avoid use on shared or third-party machines.
A compromised or misleading executable could read WeChat data or otherwise affect the local system.
The setup path recommends directly running a compiled external executable from a website or cloud drive, with no hash, signature, source pinning, or trust boundary, and that executable is expected to access the WeChat process and decrypt data.
方案A：下载 WeChatMsg.exe ... 下载编译好的 exe，直接运行 ... 它会自动找到微信进程、提取密钥、解密数据库
Avoid unverified binaries; prefer audited source builds, pinned releases, checksums/signatures, and sandboxed execution before granting access to WeChat process memory or databases.
The script can create additional local copies of sensitive WeChat message snippets in wechat_db_log.txt and wechat_search_results.json.
The included script searches message content across all message databases/tables under the chosen decrypted directory and saves snippets to local output files.
for db_path in msg_dbs: ... cursor.execute(f"SELECT message_content, create_time FROM {tbl} WHERE message_content IS NOT NULL LIMIT 1000") ... json.dump(results, f, ensure_ascii=False, indent=2)Run it only on intended databases, keep outputs in a protected location, delete logs/results when no longer needed, and consider adding chat/contact filters before scanning.
Private or third-party chat-derived details could be carried into future agent sessions or workflows beyond the original extraction task.
The artifact proposes persisting chat-derived analysis into agent memory and reusing it for ongoing monitoring, without retention limits, consent boundaries, or safeguards against sensitive context being reused in later tasks.
下一步：把这个分析结果导入MEMORY.md ... 开发一个自动化脚本：监控解密数据库中的新消息，提取关键词，实时推送需求
Do not import chat contents or summaries into persistent memory unless the data is yours to store, scoped to a clear purpose, and has an explicit retention/deletion plan.
A user may be encouraged to mine other people's chats or group conversations for business intelligence without adequate consent or legal review.
The skill frames decrypted chat analysis as useful for competitive intelligence and customer-demand mining, which can normalize privacy-invasive use despite a final legal-use disclaimer.
使用场景 ... 对手情报收集（竞品动态） ... 客户需求挖掘（AI接单方向选择）
Limit messaging-data analysis to personal backups or explicitly authorized datasets, and remove business-intelligence examples that imply collecting or exploiting others' private communications.