This release includes breaking changes for platform teams planning a safe upgrade.
✓ No known CVEs patched in this version
Topics
+5 more
Summary
AI summarySession progression now emphasizes runtime payoffs and removes clarification gating.
Full changelog
Session progression improvements based on real agent testing feedback.
Improved
- Session 2+ pitched by runtime payoff, not checklist. After Session 1, agents are told what Session 2 adds to brand-runtime.json ("agents will reject off-brand layouts and enforce visual consistency") rather than just "run brand_deepen_identity."
- Clarification gate removed. Clarification items no longer block Session 2. The conversation guide presents both in parallel — resolve clarifications AND consider visual rules — instead of gating one behind the other.
- brand_write gap surfacing. When content is generated with only Session 1 data, warnings explain what the runtime is missing concretely ("agents would know not just the right colors but how to use them").
- Report session timeline. Each session description in the HTML report now references the specific artifact it adds to brand-runtime.json.
The core insight from testing: agents stop when the task is done, not when the product wants them to continue. Sessions 2-4 need to feel like they unlock something concrete, not complete a checklist.
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 Brand-System/brandsystem-mcp
Make your brand machine-readable. Extract brand identity (colors, fonts, logo, voice, visual rules) from any website via static CSS + rendered-page extraction, compile into DTCG tokens, brand runtime contracts, and interaction policies. 34 tools across 4 progressive sessions. Subscribable `brand://runtime` and `brand://policy` MCP resources. Content compliance scoring (0-100), pass/fail gate, and HTML/CSS preflight. Brandcode Studio connector for hosted brand sync.
Beta — feedback welcome: [email protected]