Release v0.1.0-rc7

[WaylandYang]

Six days in which the typed graph became a view that keeps itself honest: alignment learned to reconcile what it writes, to shortlist what it asks about and to read a model's answer in whatever shape it comes; an errata agent now needs two votes and a structural reason before it retracts anything; a name meets its entity through a vector as well as a string; and a source can push statements straight into the open contract.

Alignment writes a view, and the view reconciles itself

• A phrase decision records the inputs it considered, so a binding is stale when its basis changed, not when a clock says so (0053, A phrase decision records the inputs it considered, so staleness is a changed basis rather than a clock #878). A human decision commits with its own recomputation job and answers 202 (0051, A human phrase decision commits with its own recomputation job and answers 202 #876).
• Materialization reconciles the rows it writes along their uniqueness timelines, as the write path always did, so a later observation of a functional attribute closes the earlier one without a manual reconcile; a document version keeps the date it was pushed with, so a same-identity update no longer makes two values look simultaneous (Materialization reconciles the rows it writes, and a version keeps the date it was pushed with #911, from the [Architecture exploration] Utopia + Jev for embodied intelligence: temporal knowledge and closed-loop decisions #875 replay in Verify derived plan steps follow changes to their premises #898).
• A shape of statement can imply a fact of another property, read once per phrase and cached (A shape of statement implies a fact of another property, read once per phrase and cached #882); a business rule may join two entities and conclude a relation (0047, Let business rules join entities and conclude relations #861); a rule's definition has a history, and a conclusion names the version it was drawn under (0060, A rule's definition has a history, and a conclusion names the version it was drawn under #913).
• The aligner reads the kind-word and phrase replies in the shapes models actually write, and asks again a bounded number of times, so a whole base no longer stays untyped in silence (Read the kind-word binding reply in the shapes the model actually writes so a whole base no longer stays untyped in silence #894, Read a phrase reply in whatever shape the model wrote it, and ask again a bounded number of times #895). When the structure fits more than ten properties it shows the ten nearest by the ontology's own vectors, describes the candidates once per batch, accepts a label where a key was asked for, and asks for rules only about bound signatures (Errata retracts on two votes and a flag, alignment shortlists its candidates, and each task thinks as hard as it needs #906).
• Each task thinks as hard as it needs: chat_reasoning_effort in the workspace settings rides on every chat call, extraction and errata run at minimal, alignment and rule proposals at low, all measured on the typed-graph bench (Errata retracts on two votes and a flag, alignment shortlists its candidates, and each task thinks as hard as it needs #906, migration 0075).

An errata agent, on two votes and a flag

An agent reviews each document's typed facts through the gate, structure first (0044 decision 7, #885). The first measured runs showed it retracting mostly what the document did say, so a retraction or revision now needs a structural flag and a second vote asked alone with the document; an unflagged fact is never retracted by the agent alone. Additions opened the other way: a name the document contains becomes an entity, and a document with no typed facts is reviewed once with an empty list (#906). On 100 documents: 34 wrong retractions to zero, recall 9.3% to 18.4%.

A name meets its entity

Recall proposes a pair by name vector, so a short form or a name in another script meets its entity through a question rather than a silent second entity (0041 cut 2, #877, migration 0080). A pair proposed by similarity rather than by the same string is never merged on the batch verdict alone: it takes the tool-using second look, or goes to a person when that look cannot run (#889). Review paging looks the two entities up before their same-name peers (#888).

A source may push statements

A Statements source accepts the open extraction contract over HTTP with its own push token: what a client sends is what the graph reads, no model in between, and it reaches the typed layer the same way a document does, through alignment. Same identity is an update, deleted is a tombstone, a key outside the contract or a subject not listed is refused at the door (0054, #884).

What a call to the endpoint survives

• An endpoint that refuses our completion ceiling has stated its own; the call is retried once at that number and the client remembers it, so gpt-4o-mini works again without a setting (Retry at the ceiling the endpoint states when it refuses ours, and remember it #914, fixes Graph extraction fails with OpenAI models in 0.1.0-rc6 (max_tokens too large) #891 from rc6).
• Two vector index builds, or a drop racing a build, take one lock instead of deadlocking (Serialize vector index builds so two concurrent CREATE INDEX CONCURRENTLY cannot deadlock #887, Take the build lock when dropping a vector index so a drop cannot deadlock a concurrent build #908); a loaded embedding runner cannot fail the elapsed-time gate (Loosen the embedding gate's elapsed-time bound so a loaded runner cannot fail it #881).
• A base is governed by one run at a time: nine overlapping governance runs were deciding the same clusters ten times over, a third of one bench run's tokens; a run that finds the base busy leaves and books one for a minute later (Govern a base with one run at a time, so concurrent runs stop deciding the same cluster repeatedly #916).

Chat and the interface

• A turn ends in exactly one earned terminal, whatever the producer's exit path, and the stream is parsed by a standard SSE parser (fix(chat): emit one earned terminal per turn #903, Assert a chat turn ends in exactly one earned terminal #858, Preserve exact chat outcomes across stream closure and reattachment #851, Keep older chat handles from unregistering newer generations #848); fallback retrieval failures stay distinct from empty results, and a failed save is never reported as a completion (Keep fallback retrieval failures distinct from empty results #850, Do not report fallback chat completion when saving fails #852, Hand off exhausted tool runs to an evidence-only final answer #845).
• Older conversations are reachable with bounded pagination, history refreshes once after an idle reattachment, stale navigation callbacks stay out of the current view (Make older chat conversations reachable through existing pagination #855, Refresh persisted chat history once after an idle reattachment #856, Keep stale chat navigation callbacks out of the current view #854).
• The chat's search_docs tool now finds the MCP guide, with a test that keeps the docs corpus and the Docs page in step (fix(server): index mcp.md in the chat-searchable docs corpus #905).

Export and provenance

• Provenance references stay inside the knowledge base (0048, fix(store): enforce same-KB provenance references #832); an export uses one snapshot and validates local references (fix(export): use one snapshot and validate local references #833); the preflight coverage follows the schema (fix(export): keep preflight coverage aligned with the schema #874).
• A literal object resolves value before summary, a typing conclusion exports its class, and prose is always xsd:string (fix(rdf): resolve object_value text robustly; emit no rdf:object when literal cannot be parsed (#831) #865); exported rules say which family they belong to (Export says which family a rule belongs to, and an axiom rule its kind and declaring predicate #910, Document the scope of exported rule identities #904).
• Document content and versions are served as a read contract (0052, feat(api): serve retained document content and versions #860).

Measured

A typed-graph bench on Re-DocRED measures what alignment recovers against the record's thresholds, waits on the right jobs, scores what errata adds and logs usage once per reply (#880, #893); the second measured runs and every cut above are tabled in scripts/bench/README.md. The #875 replay (premise-linked plan steps) runs as store tests (#898, #897, #896).

Housekeeping

Login takes the same time whether the email or the password is wrong (#867). Worker concurrency falls back to the migration's default (#871). CI runs lint, test and build as parallel jobs behind one aggregator, database-backed tests included, and the store integration tests compile as one binary (#879, #872, #873, #869, #868, #886).

Upgrading

Migrations 0070 to 0076, 0080 and 0090, all forward-only, applied on startup. An rc6 database with data was migrated and read back by this build without warnings.