From 59ccea3c1a8d3d70e03c29c1cf9a696cd02f30f7 Mon Sep 17 00:00:00 2001 From: Sam Morrow Date: Wed, 12 Aug 2026 09:20:43 +0200 Subject: [PATCH] docs: cite display templates proposal Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 3ce9ad35-1e45-451f-98f6-802946669ba8 --- docs/next-spec-iteration-priorities.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/docs/next-spec-iteration-priorities.md b/docs/next-spec-iteration-priorities.md index e420b3e..d055ca9 100644 --- a/docs/next-spec-iteration-priorities.md +++ b/docs/next-spec-iteration-priorities.md @@ -30,7 +30,10 @@ protocol proposal rather than an experimental extension, as described in the Explore how content intended for human formatting and display can be represented separately from content intended for agents. The goal is to let clients provide good human-facing experiences without compromising the structured content -agents need. This belongs in the chartered discussion of tool-response +agents need. +[PR #7 proposes the `io.modelcontextprotocol/display-templates` experimental extension](https://github.com/modelcontextprotocol/experimental-ext-tool-annotations/pull/7) +for this purpose, covering both call-side display templates and result-side +rendered text. This belongs in the chartered discussion of tool-response annotations and human-in-the-loop requirements, while any specification change would still proceed through the SEP process.