FE-1318: Reconcile canonical documentation with current topology - #414
Conversation
Thin superseded design prescriptions to historical rationale and point active guidance at current topology. Co-authored-by: Cursor <cursoragent@cursor.com>
|
Warning This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
This stack of pull requests is managed by Graphite. Learn more about stacking. |
PR SummaryLow Risk Overview Authority and planning memory: The planning-substrate assessment now states its PLAN-replacement prescription was not adopted; Superseded design notes: Large bodies in Comparison docs: Plan bookkeeping: Reviewed by Cursor Bugbot for commit 4f5538e. Bugbot is set up for automated code reviews on this repo. Configure here. |
There was a problem hiding this comment.
Pull request overview
This PR reconciles Brunch’s canonical documentation with the current, shipped topology by demoting superseded proposals to clearly-labeled historical notes and re-pointing readers to the current sources of authority (SPEC/PLAN + co-located TOPOLOGY.md + comparison runbooks/contracts).
Changes:
- Marks the planning-substrate “replace PLAN” proposal as not adopted and documents the current authority split (SPEC/PLAN + Linear/Graphite + topology files).
- Converts multiple design notes (elicitation lenses/questions, structured-exchange collapse, session-host/daemon sketches, review sets) into historical notes with pointers to current code/topology owners.
- Clarifies comparison documentation: enumerates current execution cases, distinguishes configured E2E contracts vs retained witness evidence, and states there is no single end-to-end operator command.
Reviewed changes
Copilot reviewed 14 out of 14 changed files in this pull request and generated no comments.
Show a summary per file
| File | Description |
|---|---|
| memory/PLAN.md | Records FE-1318 as completed and removes its now-closed frontier definition and “next” entry. |
| docs/README.md | Reframes the docs index around current authority (SPEC/PLAN + TOPOLOGY.md), fixes relative links, and labels historical planning notes. |
| docs/praxis/comparison-runs.md | Updates execution-case list (adds prospect-research-workspace) and tightens E2E contract vs witness language. |
| docs/praxis/comparison-guide.md | Aligns the public comparison guide with current execution cases and E2E study posture (no single operator command). |
| docs/planning/planning-record-substrate-assessment.md | Shrinks the note to durable takeaways + explicitly “not adopted” migration direction + re-entry rule. |
| docs/design/STRUCTURED_EXCHANGE_COLLAPSE.md | Rewrites as a historical note and points to current exchange topology/schemas and SPEC decision. |
| docs/design/STRUCTURED_EXCHANGE_ANSWERING_PATHS.md | Updates framing to treat the collapse doc as historical and topology as current authority. |
| docs/design/SESSION_HOST_DECISION_CANDIDATE.md | Thins to historical context and points to current app/session/rpc/web topology. |
| docs/design/REVIEW_SETS.md | Retains only the durable mechanism rationale and redirects payload/settlement authority to code + topology. |
| docs/design/ONTOLOGY_REVIEW_PROTOCOL.md | Updates language to mark retired defaulting proposals and re-points heuristics authority to current agent references. |
| docs/design/MULTI_SESSION_DAEMON_ARCHITECTURE.md | Converts to a historical note and points to current shared-host/web UI topology. |
| docs/design/ELICITATION_QUESTIONS.md | Converts the question catalogue into a historical note and points to current schema + skill references. |
| docs/design/ELICITATION_LENSES.md | Converts the lens-runtime proposal into a historical note and points to current runtime/skills/topology. |
| docs/archive/PLAN_HISTORY.md | Adds a FE-1318 closeout entry capturing what was reconciled and what remains authoritative. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Why
Some of our internal design documentation had drifted out of sync with reality — still describing earlier proposals and superseded plans as if they were live guidance. That's a real risk: anyone (including an AI coding agent) reading those docs to understand current architecture could be misled into rebuilding something we already decided against, or missing where current guidance actually lives.
What
How to test
npm run checkpasses (includes a documentation link-validity check).