mutatePageContent wrote agent edits back by DELETING the whole Yjs fragment and
re-applying a fresh Y.Doc. Yjs is a CRDT — the editor anchors its selection to
node ids — so wiping every id made an open editor's cursor lose its anchor and
snap to the end of the document on every agent write. It was most visible on
comment anchoring (issue #152): a comment changes no text, yet the cursor jumped.
(Before commit 4201f0a3 the anchoring silently no-op'd, so the destructive write
never ran for comments — hence the regression.)
Fix: write via (y-prosemirror) — the same routine the editor
uses to sync its own edits into Yjs. It structurally diffs the new doc against
the live fragment and touches only changed nodes, preserving the ids of unchanged
ones, so the cursor stays put. This improves ALL agent write tools (text edits,
node ops, comments, replace) — minimal diff instead of full replace: less collab
noise, stable block-ids, other users' cursors no longer disrupted.
- collaboration.ts: new (sanitize -> PMNode.fromJSON against
a memoized docmost schema -> updateYFragment in one transact), keeping the
findUnstorableAttr encode diagnostic; swap the destructive write-back for it.
- package.json: y-prosemirror promoted to a direct dependency (was transitive).
- test: comment-cursor-stability.test.mjs — a Yjs RelativePosition (the cursor
anchor) survives both a sibling edit and a comment-mark anchoring (the old
full-replace tombstoned it -> null). 292 package tests green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The in-app AI chat hardcoded type='page' and the shared createComment
swallowed anchoring failures silently, so agent comments never got a
text anchor/highlight.
- Forbid page-type comments for the agent: top-level comments are always
inline and require an exact `selection`; replies inherit the parent
anchor (stored as the historical `page` type).
- Throw and roll back the just-created comment when the selection cannot
be anchored, instead of leaving an orphan unanchored comment.
- Add comment-anchor module: text normalization (smart quotes, dashes,
nbsp, collapsed whitespace) and matching across adjacent text nodes
within a block, so selections crossing inline-code/bold/link anchor.
- Update create_comment (MCP) and createComment (ai-chat) tool schemas
and descriptions; add unit + mock-HTTP orchestration tests.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Implements two architecture follow-ups from the multi-aspect review.
1. Shared, zod-agnostic tool-spec registry (packages/mcp/src/tool-specs.ts)
for the 14 AI tools whose name + schema + model-facing description are
genuinely identical across the standalone MCP server and the in-app
AI-SDK chat. Both layers consume it (registerShared in index.ts;
sharedTool in ai-chat-tools.service.ts) and keep their own execute/auth.
- Zod-agnostic builders (z) => ZodRawShape bridge the zod v3 (mcp) vs
zod v4 (server) split; the registry imports no zod.
- Folds in the documented edit_page_text drift-bug fix: the stale
"strip-and-retry tolerated" claim is gone; canonical wording states a
formatting-only change is refused into failed[].
- Sibling-tool references in shared descriptions are transport-neutral so
one description is correct for both snake_case (MCP) and camelCase
(in-app) tool names.
- Loader fail-fast guard for a stale @docmost/mcp build.
- The ~17 intentionally-divergent tools (security guardrails, tuned UX)
stay per-layer, untouched.
- Rebuilt committed mcp artifacts (also regenerates a previously stale
build/lib/docmost-schema.js to match its already-committed source).
2. Formalize apps/server/test/integration/db.ts as the canonical
integration-test seed factory (module doc + a shortId helper); the
hand-written minimal seeders are kept on purpose, decoupled from the
app service-layer side effects.
Verified: server tsc + lint clean, mcp build clean; mcp unit tests 261 pass,
ai-chat-tools.service 16 pass, public-share-chat-tools 8 pass, ai-chat suite
224 pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
First, safe step of docs/backlog/ai-chat-tool-definitions-duplicated.md: the
"node may be a JSON object OR a JSON string" quirk was hand-copied at 6 tool
sites. Extract it into a single parseNodeArg() helper per package and call it at
every site. Behavior-preserving — each site's throw message is byte-identical
(patch/insert: 'node was a string but not valid JSON'; update_page_json: 'content
was a string but not valid JSON'); no tool name/description/schema changed.
Two helper copies (packages/mcp/src/lib/parse-node-arg.ts and
apps/server/src/core/ai-chat/tools/parse-node-arg.ts) are intentional: the
ESM-only @docmost/mcp cannot be imported by the CommonJS server (it is loaded at
runtime via the Function('import()') trick), so runtime code cannot cross that
boundary by a normal import. Each copy is now the single source within its
package (6 inline copies -> 2 helpers). packages/mcp/build rebuilt in sync.
Tests: parse-node-arg.spec.ts (server, Jest) + parse-node-arg.test.mjs (mcp,
node:test) — object passthrough, valid-string parse, invalid-string throw with
the right message. Server tsc clean; mcp suite 254 pass; agent structural-edit
path verified live in-browser (agent inserted a node, persisted to the doc).
Deferred (documented for the record, since the backlog doc is removed with this
commit): the FULL transport-agnostic tool-spec registry (one name+schema+
description per tool shared by both transports) and deriving DocmostClientLike
from the real client type. Both are blocked by the current architecture, not by
effort: (1) @docmost/mcp ships no type declarations and is ESM-only, so a
type-only derivation needs declaration emission + tsconfig path wiring, and the
real client's precise return types break the in-app tool test stubs (attempted,
reverted to keep tsc green); (2) the two transports intentionally DIVERGE in tool
NAMES (snake_case x38 vs camelCase x41), membership (in-app adds getCurrentPage/
listSidebarPages, omits delete_comment/image tools) and model-facing
DESCRIPTIONS, so a unified registry would change behavior on BOTH the agent and
external MCP clients and needs its own verification pass. This is forward-looking
debt (the code is correct today), to be done incrementally.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Release-cycle red-team found two same-id footnoteDefinition nodes (trivially
produced by markdown import [^d]: first / [^d]: second, or paste/duplicate)
caused silent data loss: scan() used a last-wins Map and the sync rebuild
(addToHistory:false, propagated via Yjs, un-undoable) dropped all but the last.
Fix resolves collisions so BOTH survive, with a DETERMINISTIC id scheme so
collaborators converge:
- deriveFootnoteId(originalId, occurrence, taken): the k-th (k>=2) occurrence of
id X becomes X__k, bumped with a deterministic alpha suffix only against the
doc's own id set — a pure function of document state. No Math.random/Date.now
on the sync or import paths (random uuid stays only in setFootnote, where a
single user originates a brand-new id).
- footnote-sync.resolveCollisions walks refs+defs in document order, re-ids
duplicate references via setNodeMarkup and pairs them 1:1 with definitions;
single SYNC_META-tagged transaction, returns null when canonical (terminates).
- Markdown import (footnote.marked) + MCP mirror (collaboration.ts) dedup with
the same deterministic scheme + marker rewrite; packages/mcp/build regenerated.
- Paste plugin remaps colliding pasted ids against the current doc.
Tests: two independent editors resolving the same duplicate-id doc produce
IDENTICAL ids (the cross-client determinism guard that the random version would
fail); both definitions survive the first edit; import dedup is deterministic.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds footnotes: a superscript marker in the text linked to an editable
definition in a Footnotes section at the end of the page, with auto-numbering
and a read-only hover popover. Chose the reference+definitions model (3 plain
nodes) over an inline atom with a sub-editor specifically for collaboration
safety.
editor-ext (packages/editor-ext/src/lib/footnote/):
- footnoteReference (inline atom, id), footnotesList (block, last child),
footnoteDefinition (paragraph+, id). renderHTML emits sup[data-footnote-ref]
/ section[data-footnotes] / div[data-footnote-def]; parse-rule priority makes
the empty reference win over the Superscript mark (else it is dropped on the
server save).
- numbering: a decoration-only plugin (pure function of doc order) -> every
client computes identical numbers, no document mutation, Yjs-safe.
- sync plugin: single-pass, always SYNC_META-tagged and skipping remote txns
(terminates, no loop), idempotent; canonicalizes to one trailing footnotesList
(merging duplicates), creates missing definitions, drops orphans, and
coexists with TrailingNode. Disabled in read-only.
- commands setFootnote (one tx: reference + definition at the matching index +
focus) / removeFootnote (cascade, one undo) / scrollTo*. slash /footnote.
client: superscript NodeView + floating-ui read-only popover; bottom-list and
definition NodeViews; registered in mainExtensions.
server: the three nodes registered in tiptapExtensions so collab/save/export
keep them. Round-trip regression spec guards the Superscript parse-priority.
markdown: turndown/marked round-trip to pandoc/GFM [^id] (+ a code-fence guard
so footnote-like lines inside code blocks are not extracted).
MCP mirror: schema + markdown-converter + commentsToFootnotes rewritten to real
footnote nodes + diff marker counting; NUL sentinels written as \u0000 escapes.
v2 follow-ups (per plan): definition reordering on reference move, id-collision
regeneration on paste, multiple references to one footnote.
Implements docs/footnotes-plan.md (variant B).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
list_pages gains an opt-in `tree` parameter on both surfaces (the
@docmost/mcp server tool and the AI-chat agent tool), which share the
same DocmostClient.listPages. Default behavior (recent-by-updatedAt flat
list) is unchanged.
- client.ts: listPages(spaceId?, limit=50, tree=false); when tree is
true it requires spaceId (throws a specific error otherwise), walks the
sidebar tree via the existing bounded/cycle-safe enumerateSpacePages,
and returns a nested tree; limit is ignored in tree mode.
- lib/tree.ts: new pure buildPageTree() — lean nodes { id, slugId, title,
children? }, children sorted by position (code-unit order), orphans
promoted to roots, cycle-safe.
- index.ts + ai-chat-tools.service.ts: expose `tree` in the tool schemas
and descriptions; docmost-client.loader.ts: mirror the new signature.
- tests: add packages/mcp/test/unit/tree.test.mjs (nesting, ordering,
lean shape, orphan promotion, cycle/self-reference safety).
- rebuild @docmost/mcp (build/ is tracked and loaded at runtime).
edit_page_text reported "success" when asked to change formatting (e.g. remove
strikethrough): the markdown-strip fallback matched the bare text, the replace
preserved marks, and the tool returned success — so the agent believed it had
fixed something that never changed.
Two fixes, both in the shared @docmost/mcp DocmostClient so they reach BOTH the
standalone MCP server and the in-app AI chat (which loads @docmost/mcp):
- Verifiable result for every content mutator: mutatePageContent now computes a
`verify` change-report (text inserted/deleted, blocks changed, per-mark-type
delta, integrity/structure delta) via summarizeChange() and returns it on all
mutators (incl. replaceImage via mutateLiveContentUnlocked). diffDocs is
text-only, so the mark/structure delta is what surfaces formatting changes.
- edit_page_text hard-refuses formatting edits: applyTextEdits rejects an edit
whose find/replace differ only in markdown markers (via stripBalancedWrappers,
which strips balanced wrappers/links without trimming whitespace/emoji, so
plain-text edits like trailing-space trims, snake_case, math are NOT refused).
A fully-refused batch errors instead of silently succeeding.
Also updated the model-facing edit_page_text descriptions in BOTH tool layers
(packages/mcp/src/index.ts and ai-chat-tools.service.ts) to drop the misleading
"strip-and-retry tolerated" wording and point formatting changes to patch_node.
New unit tests: test/unit/diff-verify.test.mjs, test/unit/json-edit-refuse.test.mjs.
Locators (edit_page_text `find`, insert_node `anchorText`) are matched
against the document's plain text, so a model-supplied locator carrying
markdown wrappers (**bold**, *italic*, `code`, [t](url)) or trailing emoji
never matched and the edit/insert failed. Add stripInlineMarkdown() and a
fallback: try the locator verbatim first (exact match wins, so literal
asterisks/underscores still work), and only on zero matches retry with a
markdown-stripped form. The ambiguity guard runs on the post-fallback count,
and `replace` / inserted node content are never stripped, so no formatting is
lost. Failed edits gain an atom-aware reason plus a bounded "closest block
text" hint; the insert_node "anchor not found" error now points at plain-text
anchors / anchorNodeId.
New packages/mcp/src/lib/text-normalize.ts (+ unit tests); wired into
json-edit.ts and node-ops.ts; tool descriptions updated. Tests: 212 pass.
edit_page_text (applyTextEdits) now matches at the inline-block level instead of
per text node, so a find/replace may cross bold/italic/link boundaries; the
replacement inherits marks from the unchanged common prefix/suffix via a diff
splice. Atom (non-text inline) slots can never be part of a match, making the
U+FFFC placeholder collision-safe, and inserted text never inherits an atom's
marks.
The edit batch is no longer all-or-nothing: applyTextEdits returns
{ doc, results, failed } and applies what it can; editPageText writes only on a
real change (no spurious history version for a no-op) and throws an aggregated,
actionable error only when nothing applied.
The AI-chat insert_node / patch_node / update_page_json tools now JSON.parse a
node/content argument that arrives as a string, matching the standalone MCP
server (this is what made insert_node fail under OpenAI tool calls).
Tool descriptions gain concrete ProseMirror examples and reflect the new
edit_page_text behavior. Adds/updates json-edit unit tests (183 pass).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the removed enterprise EE MCP (private apps/server/src/ee submodule,
license-gated /mcp route) with our docmost-mcp, vendored as an isolated ESM
workspace package and served by the server over HTTP — no enterprise license.
Backend:
- Add packages/mcp (@docmost/mcp): vendored docmost-mcp refactored into a
side-effect-free createDocmostMcpServer() factory (38 tools preserved),
stdio entry kept in stdio.ts, Streamable-HTTP session manager in http.ts.
- Add apps/server McpModule: @Post/@Get/@Delete('mcp') (served at /mcp via the
existing global-prefix exclude), @SkipTransform + reply.hijack to bridge raw
Fastify req/res into the SDK transport. The module dynamically imports the
ESM-only package from CommonJS via a Function-indirected import resolved with
require.resolve + file:// URL. Gated by the workspace ai.mcp toggle, a
service-account (MCP_DOCMOST_EMAIL/PASSWORD/API_URL) and optional MCP_TOKEN;
per-session idle eviction (MCP_SESSION_IDLE_MS).
- Drop the enterprise license check on mcpEnabled in workspace.service.
- Dockerfile: copy packages/mcp into the production image.
- .env.example: document MCP_DOCMOST_*, MCP_TOKEN, MCP_SESSION_IDLE_MS.
Frontend:
- Recreate the community "AI & MCP" workspace-settings panel (mcp-settings.tsx):
admin-only toggle on settings.ai.mcp with optimistic update, copyable
${APP_URL}/mcp URL; wired into workspace-settings page. Reuses existing i18n.
Fixes:
- Pin packages/mcp tiptap deps to 3.20.4 (matching the client) and inline
getStyleProperty, preventing a duplicate @tiptap/core@3.26.1 from leaking into
the client editor via pnpm shamefully-hoist (was breaking apps/client tsc).
