How to Contribute to the OpenClaw Threat Model
This page explains how anyone can contribute to the OpenClaw threat model by adding threats or suggesting mitigations. It is intended for community members and security researchers.
Read this when
- You want to contribute security findings or threat scenarios
- Reviewing or updating the threat model
The threat model is a document that evolves over time. Anyone can contribute; you do not need a background in security or MITRE ATLAS.
Note
This process is for expanding the threat model, not for reporting active vulnerabilities. If you have found an exploitable vulnerability, follow the responsible disclosure guidelines on the Trust page instead.
Ways to contribute
Add a threat. File an issue on openclaw/trust and describe the attack scenario in your own words. The following are helpful but not mandatory:
- The attack scenario and how an attacker could exploit it.
- Which components are affected (CLI, gateway, channels, ClawHub, MCP servers, and so on).
- Your assessment of severity (low, medium, high, or critical).
- References to related research, CVEs, or real-world examples.
During review, maintainers assign the ATLAS mapping, threat ID, and risk level.
Suggest a mitigation. Open an issue or a PR that references the threat. Be concrete and actionable: "per-sender rate limiting of 10 messages per minute at the gateway" is far more useful than "implement rate limiting."
Propose an attack chain. Attack chains illustrate how several threats combine into a realistic scenario. Outline the steps and how an attacker would link them together; a short narrative is better than a formal template.
Fix or improve existing content. Typos, clarifications, outdated information, better examples: PRs are welcome and do not require an issue.
Framework reference
Threats are mapped to MITRE ATLAS (Adversarial Threat Landscape for AI Systems), a framework for AI and ML specific threats such as prompt injection, tool misuse, and agent exploitation. You do not need to know ATLAS to contribute; maintainers handle the mapping during review.
Threat IDs. Each threat receives an ID like T-EXEC-003, assigned by maintainers during review.
| Code | Category |
|---|---|
| RECON | Reconnaissance - information gathering |
| ACCESS | Initial access - gaining entry |
| EXEC | Execution - running malicious actions |
| PERSIST | Persistence - maintaining access |
| EVADE | Defense evasion - avoiding detection |
| DISC | Discovery - learning about the environment |
| EXFIL | Exfiltration - stealing data |
| IMPACT | Impact - damage or disruption |
Risk levels. If you are unsure about the level, simply describe the impact; maintainers will assess it.
| Level | Meaning |
|---|---|
| Critical | Full system compromise, or high likelihood combined with critical impact |
| High | Significant damage likely, or medium likelihood combined with critical impact |
| Medium | Moderate risk, or low likelihood combined with high impact |
| Low | Unlikely and limited impact |
Review process
- Triage - new submissions are reviewed within 48 hours.
- Assessment - maintainers verify feasibility, assign ATLAS mapping and threat ID, and validate the risk level.
- Documentation - formatting and completeness are checked.
- Merge - the submission is added to the threat model and visualization.
Resources
Contact
- Security vulnerabilities: Trust page for reporting instructions, or
security@openclaw.ai. - Threat model questions: file an issue on openclaw/trust.
- General chat: Discord
#securitychannel.
Recognition
Contributors to the threat model are acknowledged in the threat model acknowledgments, release notes, and the OpenClaw security hall of fame for significant contributions.