Add second-pass chat UX milestones

Extend the chat-ergonomics roadmap with the remaining UX work highlighted by the readiness review.

Document a second pass focused on removing shell glue from canonical CLI handoff flows, making the recommended chat-host profile more obvious without changing 3.x compatibility defaults, and polishing human-mode content reads for cleaner transcripts and copy-paste behavior.

Keep these milestones explicitly workspace-first and scoped to product UX, with CLI-only shortcuts allowed where the SDK and MCP surfaces already provide the structured behavior natively.

docs/roadmap/llm-chat-ergonomics.md

@@ -33,13 +33,24 @@ More concretely, the model should not need to:
 - choose from an unnecessarily large tool surface when a smaller profile would
   work
 
+The remaining UX friction for a technically strong new user is now narrower:
+
+- the best chat-host profile is recommended in docs, but not yet obvious enough
+  from the default live `mcp serve` path
+- canonical CLI walkthroughs still need small amounts of shell glue such as
+  `python -c` extraction of `workspace_id` and `$(cat fix.patch)` expansion
+- human-mode file reads are functional, but still need final transcript polish
+  for copy-paste and chat logs
+
 ## Locked Decisions
 
 - keep the workspace product identity central; do not drift toward CI, queue,
   or runner abstractions
 - keep disk tools secondary and do not make them the main chat-facing surface
 - prefer narrow tool profiles and structured outputs over more raw shell calls
-- every milestone below must update CLI, SDK, and MCP together
+- capability milestones should update CLI, SDK, and MCP together
+- CLI-only ergonomics are allowed when the SDK and MCP surfaces already have the
+  structured behavior natively
 - every milestone below must also update docs, help text, runnable examples,
   and at least one real smoke scenario
 
@@ -50,6 +61,9 @@ More concretely, the model should not need to:
 3. [`3.4.0` Tool Profiles And Canonical Chat Flows](llm-chat-ergonomics/3.4.0-tool-profiles-and-canonical-chat-flows.md) - Done
 4. [`3.5.0` Chat-Friendly Shell Output](llm-chat-ergonomics/3.5.0-chat-friendly-shell-output.md) - Done
 5. [`3.6.0` Use-Case Recipes And Smoke Packs](llm-chat-ergonomics/3.6.0-use-case-recipes-and-smoke-packs.md) - Done
+6. [`3.7.0` Handoff Shortcuts And File Input Sources](llm-chat-ergonomics/3.7.0-handoff-shortcuts-and-file-input-sources.md) - Planned
+7. [`3.8.0` Chat-Host Onramp And Recommended Defaults](llm-chat-ergonomics/3.8.0-chat-host-onramp-and-recommended-defaults.md) - Planned
+8. [`3.9.0` Content-Only Reads And Human Output Polish](llm-chat-ergonomics/3.9.0-content-only-reads-and-human-output-polish.md) - Planned
 
 Completed so far:
 
@@ -67,6 +81,15 @@ Completed so far:
   cases so the stable product is now demonstrated as repeatable end-to-end stories instead of
   only isolated feature surfaces.
 
+Planned next:
+
+- `3.7.0` removes the remaining shell glue from canonical CLI flows with shortcut flags for
+  identifier handoff and file-backed text inputs.
+- `3.8.0` makes the recommended chat-host entrypoint obvious from the top-level docs, help text,
+  and shipped MCP examples without changing the `3.x` compatibility default.
+- `3.9.0` makes human-mode file reads cleaner in terminals and chat logs, with explicit
+  content-only reads where summaries would otherwise get in the way.
+
 ## Expected Outcome
 
 After this roadmap, the product should still look like an agent workspace, not
@@ -79,4 +102,7 @@ The intended model-facing shape is:
 - file edits are structured and model-native
 - workspace discovery is human and model-friendly
 - shells are readable in chat
+- CLI handoff paths do not depend on ad hoc shell parsing
+- the recommended chat-host profile is obvious from the first MCP example
+- human-mode content reads are copy-paste safe
 - the five core use cases are documented and smoke-tested end to end