AgentMD is an alpha project. Only the latest commit on the maintained branch is eligible for security fixes. Published alpha tags are not production support commitments.

Do not open a public issue for vulnerabilities, leaked credentials, or active repository compromise. Use GitHub's private vulnerability reporting for this repository. Include:

• affected commit or release
• reproduction steps
• expected impact
• relevant artifact, receipt, or workflow hashes
• whether credentials or outbound network access were involved

Do not include live secrets in the report. Revoke exposed credentials before sharing redacted evidence.

• Lead Artifacts are untrusted input.
• Trust scores are advisory and never authorize execution.
• Receipt hashes detect changes but are not signatures.
• Skill edits are limited to workspace-local skills/*/SKILL.md files.
• AgentMD does not support executable serialization formats.
• External adapters are not executed by the v0.x runtime.

See docs/threat_model.md, docs/receipt_integrity.md, and docs/security_hardening_todo.md for current guarantees and open work.