Only the latest minor version of money-desk receives security fixes.
At this stage of the project that is the 0.19.x series.
| Version | Supported |
|---|---|
| 0.19.x | ✅ |
| < 0.19 | ❌ |
Please do NOT open public GitHub issues for security vulnerabilities.
If you believe you have found a security issue in money-desk — for example:
- A prompt-injection vector inside a specialist or skill markdown that could cause the agent to ignore its guardrails (analysis-only, PII-masking, citation requirement).
- A path traversal or arbitrary-file-write in
bin/cli.mjs. - A way for the installer to escape the intended install directory.
- An injection vector in any future generated content.
- Any way Money Desk could exfiltrate user data beyond the normal Copilot CLI ↔ Copilot model service flow described in PRIVACY.md.
…please report it privately using GitHub's private vulnerability reporting feature.
Include in your report:
- A description of the issue and its impact.
- Steps to reproduce, or a minimal proof of concept.
- The affected version(s) and environment (OS, Node version, Copilot CLI version).
- Your suggested remediation if you have one.
- Acknowledgement: within 7 days.
- Initial assessment: within 14 days.
- Patch + advisory: as fast as the severity warrants. Coordinated disclosure preferred.
This project ships prompts and a thin SDK adapter — there is no server, no database, no network code, and no runtime dependencies. The supply-chain attack surface is therefore very small. The most likely class of security issue is prompt-level: a specialist or skill markdown that allows a crafted input document to override the agent's guardrails. Reports of that kind are very welcome.
For privacy concerns that are not strictly security vulnerabilities, please open a normal GitHub issue or PR — see PRIVACY.md.