Skip to content

Security: aoci-spec/aoci-code

SECURITY.md

Security Policy

Supported versions

Security fixes are evaluated for the latest maintained AOCI-CODE version and the current default branch. Older development snapshots are not maintained as separate security lines unless a release notice says otherwise.

Reporting a vulnerability

Use GitHub private vulnerability reporting for the canonical repository:

https://github.com/aoci-spec/aoci-code/security/advisories/new

Do not disclose an exploitable vulnerability, credential, private endpoint, proprietary source, or personal data in a public issue. If private reporting is temporarily unavailable, open a public issue containing only a request for a confidential contact; do not include sensitive details.

A useful report includes:

  • affected version, commit, binary SHA-256, operating system, and architecture;
  • the affected boundary, such as path containment, MCP stdout isolation, secret handling, CAS, atomic writes, archive extraction, or update verification;
  • minimal reproduction steps and expected impact;
  • whether the issue is public or actively exploited;
  • evidence with secrets and third-party source removed.

Maintainers will acknowledge the report, assess affected versions, coordinate remediation and disclosure, and publish guidance when disclosure is safe.

Security boundaries

  • MCP stdio stdout is reserved for JSON-RPC; logs and diagnostics use stderr.
  • Configuration stores credential environment-variable names, not values.
  • Optional model access uses only an explicitly configured endpoint.
  • Formal asset updates use validation, locking, CAS, atomic writes, and recovery.
  • Database Evidence collectors are limited to schema/catalog metadata and do not issue DDL or DML.
  • Checksums establish artifact integrity but do not identify the publisher; release provenance and signing are separate controls.

Dependency and vulnerability scan findings require reachability and impact review. Suppressions must be narrow, documented, and reviewed.

There aren't any published security advisories