Security audit
investment-framework
Security checks for vulnerabilities and agentic risk
Overview
The skill largely matches an investment-analysis purpose, but the repository contains deployment/publish scripts and runnable shell tooling (cron, auto-publish, token hints) that are unrelated to analysis and raise risk of credential exposure or arbitrary command execution.
This skill appears to contain a legitimate investment-analysis framework, but review the repository before installing or granting runtime execution privileges. Specific checks to do: (1) Search scripts for hard-coded tokens or API keys (auto-publish-clawhub.sh, CLAWHUB_* docs mention a token). Remove or rotate any embedded secrets. (2) If you do not need publishing/cron behavior, do not run auto-publish or crontab scripts and consider removing those files. (3) Limit allowed tools: prefer disabling Exec/Bash/Write unless you trust the code and run it in an isolated sandbox. (4) Confirm where API keys are expected — require explicit environment variables rather than in-repo storage and verify the skill declares them. (5) Audit any scripts that modify system state (crontab, /tmp write, uploads) and network endpoints they contact (eastmoney, qt.gtimg.cn, alphavantage, clawhub). (6) If you must run it, run in an isolated VM/container and review logs/network traffic; do not run on a machine containing sensitive credentials. If you want, I can (A) search the repository for hard-coded tokens/keywords, (B) list files that would write to crontab or /tmp, or (C) produce a minimal configuration that disables publish scripts and limits tool permissions.
SkillSpector findings are pending for this release.