This release includes 1 security fix for security teams reviewing exposed deployments.
Topics
+8 more
Affected surfaces
Summary
AI summaryCross-model continuity lets tasks, memory, and governance persist across Claude Code, Codex, and Gemini CLI.
Full changelog
What's New
Cross-model continuity — Your tasks, memory, and governance carry between Claude Code, Codex, and Gemini CLI. Switch models without losing context.
Project scan — delimit_scan discovers OpenAPI specs, frameworks, security issues, and tests in your project. The first-run "aha moment."
CLI-first deliberation — Uses your existing CLI subscriptions (Claude Pro, Codex Pro, Gemini Ultra) instead of separate API keys. No extra cost.
Governance trigger — AI assistants automatically check the ledger on session start. Works across Claude Code, Codex, Cursor, and Gemini CLI.
Security hardening:
- Pinned Python dependencies with isolated venv
- Fixed command injection in test_smoke (shell=True removed)
- All internal paths removed from shipped package
- Input validation on subprocess calls
GitHub Action v1.5.2:
- Enforce mode now posts PR comment before failing CI
- YAML detection fix for base branch specs
- Live demo with 6 breaking changes
81 tools including: scan, release_sync, models configure, deploy_npm
Install
npx delimit-cli setup
No API keys. No account. 10 seconds. Works with Claude Code, Codex, Cursor, and Gemini CLI.
Full Changelog: https://github.com/delimit-ai/delimit/commits/main
Security Fixes
- Fixed command injection in `test_smoke` by removing `shell=True` usage.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About delimit-ai/delimit
API governance server that detects breaking changes in OpenAPI specs. Diffs two spec versions, applies configurable policy rules (strict/default/relaxed), and returns structured pass/fail verdicts. 23 change types, 10 breaking. Supports OpenAPI 3.0, 3.1, and Swagger 2.0.
Related context
Beta — feedback welcome: [email protected]