Status
What works today, what is in development and what is planned, with the evidence for each claim. Terms such as Slice and Candidate are explained in The words.
Where Factory stands
As of 4 October 2026. Each line is read again from its source whenever this site is built, and the build stops if the source no longer says it.
-
Bounded local development is complete: 36 installed module families and 2 full 13-stage workflow rehearsals independently qualified on 4 October 2026. The accepted source passed 3321 root and 27 Dashboard checks. Current scope and merged-main CI.
Source:
docs/factory-completion-joint-qualification-20261004-receipt.json -
Factory is accepted local infrastructure in development, on one Mac.
Source:
docs/agents/factory-model.md -
No Factory product release, production service or customer is established. The marketing and documentation sites are published.
Source:
docs/agents/factory-model.md -
All 42 epics are open: 27 customer-value epics and 15 adoption epics. Each requires its own independent evidence to close.
Source:
docs/epics.json -
No model route is qualified yet, so no model runs inside Factory: both producer receipts say unavailable.
Source:
docs/claude-producer-receipt.json·docs/openai-producer-receipt.json -
The joined delivery stages passed independent bounded local qualification. Live providers, complete external Workflow adoption, release and customer benefit remain unestablished.
Source:
docs/build-review.md·docs/agents/delivery.md -
Six local products are registered for read-only metadata inspection. Registration grants no right to change, release or spend.
Source:
docs/onboarding-receipt.json -
Every module accepted before 1 October 2026 was reviewed independently by OpenAI's gpt-6-astra before acceptance, and failed rounds are recorded: Producers passed on its ninth round, Repository Assurance and the Product gate on their sixth.
Source:
docs/build-review.md -
On 1 October 2026 the coordinating Claude Code session, as the owner's acceptance reviewer, accepted 24 module families of the merged adoption work locally, each with its own receipt, after independent Claude Opus 5.5 reviews and an advisory Fable 5.1 review of each. The acceptance covers local infrastructure only, not release or customer value.
Source:
docs/build-review.md
What the labels mean
- Working today · verified
- It runs on the owner's Mac today, and a different model's independent review accepted it. Its scope is stated beside it: usually the local fixtures or the local dashboard, never a release.
- In development
- Parts are built and independently reviewed; some still await acceptance. The joined stages passed bounded local qualification on synthetic Products. Adoption in a commissioned external system and live operation still have explicit gates.
- Planned
- Designed or directed by the owner; nothing of it is built. Where the plan is written down is linked beside it.
Working today · verified
These run on the owner's Mac today. Each states its scope.
-
Record every decision once
Working today · verified
Each accepted Command is written once, with its result and the Events it caused. Repeating a command returns the recorded answer, so the decision is never made twice; a changed command under the same ID is refused.
Scope: One local SQLite file on one Mac, crash-tested by killing processes; power loss is not tested. Anything a caller does outside the journal can still happen twice.
Evidence: Command journal guide · Existing-store receipt (JSON) · Inspection receipt (JSON) · Build review
-
Check a fixture and give an honest Verdict
Working today · verified
A built-in test fixture runs as a real, bounded process. What it did is stored as Evidence named by its own contents, and Assurance gives a Verdict: Verified, Failed or Inconclusive. Anything missing, altered or unclear never counts as a pass.
Scope: The four built-in test fixtures only (pass, fail, hang and noisy), run from the command line as the local development reference shows. Not a product's code.
Evidence: Evidence and Assurance guide · Fixture worker guide · Baseline receipt (JSON) · Build review
-
Recover from a controller crash within Budget
Working today · verified
Each Attempt has exactly one owner and a private workspace. If the controller dies partway, the next run settles that Attempt from its stored report, or proves the old work cannot resume, and starts at most a bounded replacement. A Run's Budget is never extended.
Scope: The built-in fixtures on one Mac. Recovery happens when someone runs
verifyorrecover, never in the background.Evidence: Guarded verification guide · Attempt fence guide · Recovery receipt (JSON) · Build review
-
See each product's sourced progress, with unknown shown as unknown
Working today · verified
A local dashboard shows what sources report for each registered product, with the source and the time. A number that is partial or unknown says so, and why; only an exact count ever shows 0.
Scope: Six local products, registered for read-only metadata inspection. Facts arrive by trusted local import, the dashboard answers only on this Mac, and customer value always shows as unknown.
Evidence: Portfolio guide · Dashboard guide · Portfolio receipt (JSON) · Dashboard receipt (JSON) · Onboarding receipt (JSON)
-
Keep each product's core values as exact revisions
Working today · verified
Up to eight values per product, each with the rule it applies when choosing between options. Every save is a new revision, and anything that cites a revision gets exactly that revision back, even after later edits.
Scope: Edited in the local dashboard and pinned by owner-selected planning in the bounded local workflow. Values grant no permission; automatic live planning remains unqualified.
Evidence: Project guidance guide · Guidance receipt (JSON)
-
One list of what needs attention, from sources only
Working today · verified
Open blockers, unavailable repositories and failing sources in one list. Each blocker has a guide: what is blocked, who can act, what was tried and the next step. Missing data never becomes an action.
Scope: A blocker resolves only when its source reports it; Factory does not re-check the prerequisite itself.
Evidence: Required actions guide · Guidance receipt (JSON)
In development
Parts of these are built and independently reviewed; some still await acceptance. Each says what is still missing.
-
Deliver one useful change, end to end
In development
Join the accepted parts into one flow: a Slice admitted under your grant, a Candidate built and sealed, checked in the sandbox, judged, and integrated once, with every step recorded.
Still missing: The joined flow and command line passed independent bounded local qualification. Complete Workflow adoption and one useful change delivered to a commissioned real Product remain open; see current local evidence (JSON).
Evidence: Handover: next steps (not published) · Build review
-
Ask a model for a bounded proposal, through your subscription
In development
One model, asked once with every tool switched off, proposes small text-file edits for a Slice. Its answer is only data: Repository decides whether it can be sealed.
Still missing: No model route is qualified, so nothing is proposed yet. By the owner's decision of 1 October 2026 the Claude command line stays unavailable for Factory's model roles, and live qualification of the ChatGPT sign-in route through Codex remains open.
Evidence: Producers guide · Claude receipt: unavailable (JSON) · OpenAI receipt: unavailable (JSON)
-
Seal a proposed change as an exact Candidate
In development
Proposed edits become a Candidate: an exact commit and manifest that cannot change afterwards without being detected, exported byte for byte for checking.
Still missing: Repository sealing is exercised by the local synthetic Delivery and Implementation composition. No real Product's code is qualified by that proof.
Evidence: Repository guide · Sealing receipt (JSON)
-
Check a real change in a sandbox, then judge it
In development
Each of a Slice's frozen checks runs once in a macOS sandbox with no network and nothing writable. What the host observed becomes Evidence, and Repository Assurance judges it: Verified, Failed or Inconclusive.
Still missing: The local delivery composition passed independently reviewed synthetic checks of the sandbox, collector and judge. Complete external Delivery adoption and live Product qualification remain open; see current local evidence (JSON).
Evidence: Confined executor guide · Executor receipt (JSON) · Repository collector guide · Repository Assurance guide
-
Admit delivery work only within your grants
In development
One durable record per product holds your grants, their revocations, a stop switch and the one current owner. It freezes each Slice with its allowance and admits every build, check and branch move only within it.
Still missing: The Product gate passed bounded local qualification with Delivery, Implementation and Verification. External Workflow adoption and live Product delivery remain open.
Evidence: Product gate guide · Build review
-
Move a branch at most once, and say unknown when unsure
In development
The target branch moves from the Candidate's base to its commit at most once per prepared Action. If Factory cannot confirm the move, it reports unknown and never retries blindly; a last check just before the move can hold it back.
Still missing: Accepted for fresh local repositories. The independently qualified bounded local flow joins it to the Product gate with retained dispatch identity and a final authority guard. Commissioned real-Product delivery remains open; it contacts no remote.
Evidence: Repository guide · Integration receipt (JSON) · Last-check receipt (JSON)
-
Use a model only where it is needed
In development
One registry says which parts may use a model, which one, at what effort and why. Most never do: recording, checking and showing work must give the same answer every time. Each model task may name up to four models in priority order; the first that nothing recorded refuses is chosen, a checker never shares a model with what it checks, and the registry refuses rather than substitutes a model.
Still missing: No model is qualified, so every model task refuses today, and nothing dispatches. The priority lists are merged after independent review; acceptance is not claimed.
Evidence: Intelligence guide · Per-module table in the README (not published)
-
A short digest of what needs you, with its sources
In development
Branch health reads each product's main branch and says red, green or unknown from what GitHub reports. The owner digest turns recorded facts into short updates, with a Monday digest of at most 150 words and every source a click away, and a ledger counts what Factory asks of you and the minutes it takes.
Still missing: Each part passed independent review on recorded and test data only. Nothing reads GitHub live, runs on a schedule or sends an update, and no real ask is recorded yet: live polling waits for your commission.
Evidence: Owner digest guide · Branch health guide · Owner policy guide · Build review
-
Use Factory's parts in another system
In development
Each foundational module becomes a package another program can install and run with its own storage, clock and identities. Every package is proved by a separate program that uses it from the outside and imports nothing private, and old records keep replaying exactly as they were written.
Still missing: The packaging work merged into the main branch on 1 October 2026 after review, and its 24 module families were accepted locally that day, each with a receipt. No package is published. The 4 October independent qualification covers 36 installed families, individually callable stages and two full workflow rehearsals. Complete Workflow adoption, live providers and customer delivery remain separate; see current local evidence (JSON).
Evidence: Independent adoption · Adoption inventory · Implementation plan · Build review
-
Plan dates, capacity and timing without inventing certainty
In development
A date keeps the precision its source gave: a quarter stays a quarter, and two sources that disagree are shown as a conflict, not settled by whichever came last. A capacity range comes only from comparable reviewed work, with its assumptions stated; otherwise it is unknown. A proposed time is checked against notice periods, embargoes and clock changes.
Still missing: Three evaluators passed independent review and were accepted locally on 1 October 2026; each answers only for the data it is given. Nothing stores a plan, proposes a date, reserves capacity or starts work.
Evidence: Temporal Planning guide · Capacity planning guide · Material timing guide · Planning design
-
Find customer problems in public sources
In development
Each product's public sources are read on a schedule: a basic pass daily and a full pass weekly. Code fetches, de-duplicates and keeps an exact quote and hash; a small, fast model sorts new items into Signals.
Still missing: Two parts are built and each passed independent review, tried only on test data: the schedule, which names each nightly and weekly pass once and holds its weekly token allowance, and the reader, which fetches only the public pages you commission and keeps exact quotes with their hashes. Research jobs join attributed collection through an explicit caller contract. Live scheduling, semantic triage and synthesis remain unqualified, and the detailed admission policy still awaits your decisions (P-C20 in the project intent).
Evidence: Discovery sources guide · Lifecycle schedule guide · Discovery design · Build review · Project intent (not published)
-
Take in service failures as Signals
In development
Crashes, errors and slow requests reported by a product's services become Signals, each kept with where it came from, which product it belongs to and whether it may be kept at all. A restart or a lost reply never makes the same observation new, and a source that cannot be reached never reads as healthy.
Still missing: The intake passed independent review on test data. No live service is connected, and each will need a token you provide; Intake and Repair have bounded local evidence, but real service feedback and production recovery remain unqualified.
Evidence: Operational signals guide · Intake and Repair design · Build review
-
Plan from each product's saved values
In development
A brief pins the exact revision of a product's values it follows, applies their rules and explains the trade-offs. Values guide choices; they never grant authority or weaken acceptance.
Still missing: The bounded local workflow pins owner Guidance before selecting a plan, and a selected plan generation can become a work brief. Automatic planning and implementation for a commissioned real Product remain unqualified.
Evidence: Factory workflow guide · Work brief guide · Joint qualification (JSON)
Planned
Nothing of these is built yet. Each links where it is planned.
-
Admit, park or reject proposals, rules first
Planned
Once a week a larger model turns the strongest clusters of Signals into a few proposals. Fixed rules decide first; a second model may challenge a proposal but only ever lower it; anything that widens authority goes to you.
Plan: Owner direction of 29 September 2026 (P-C20 in the project intent).
Sources: Project intent (not published)
-
Ask you only what Factory cannot decide
Planned
A short, focused list of decisions only you can make, each with its objective, trade-offs, a recommendation and the Evidence one click away. Routine, reversible work never waits for approval.
Plan: The decision interface design and the project intent (P-C01, P-C16).
Sources: Decision interface design · Project intent (not published)
-
Release only with your approval
Planned
Release notes and launch copy are drafted from checked facts only. Every outward action (publishing, releasing, contacting customers, spending) waits for your approval of that exact action, and risky changes reach an alpha channel first.
Plan: The design's release and marketing rules, and owner direction of 29 September 2026 (P-C20).
Sources: Design: deliver with evidence · Project intent (not published)
-
Measure whether a change helped
Planned
A measurement contract is fixed before release: baseline, target, window and guardrails. When the window closes, an Assessment concludes Supported, Rejected or Inconclusive; only Supported counts as value. A/B tests are used only under genuine uncertainty.
Plan: The design's measurement rules, and owner direction of 29 September 2026 (P-C20).
Sources: Design: measurement · Project intent (not published)
-
Keep, renew or retire, and improve in bounded steps
Planned
Capabilities are kept while they serve customers and retired deliberately, with migration and notice. Nightly research and weekly simplification enter as ordinary verified Slices, and anything a clean-up breaks is restored.
Plan: The self-improvement design and next step 4 in the handover. No schedule is active.
Sources: Self-healing and self-improvement · Handover: next steps (not published)
-
Reuse the lifecycle in a second product
Planned
A second product reuses the process and its proven parts while its data, authority and behaviour stay separate.
Plan: Epic E18 in the implementation guideline.
Sources: E18 in the implementation guideline
What is not established
Not established yet, in plain words:
- No Factory product release, production service or customer. The marketing and docs sites are published; no customer benefit has been measured.
- No qualified model route, so no model runs inside Factory yet.
- No scheduled automation: nothing runs nightly or weekly, and nothing plans or starts work by itself.
- The composed workflow is independently accepted within its bounded local profile. Complete Workflow adoption in a commissioned external system remains open.
- The 24 module families accepted on 1 October 2026 retain their original evidence. The 4 October qualification covers 36 installed families and 2 full 13-stage rehearsals. Current scope and exact-main CI. Live use and release remain unestablished.
- Implementation, Verification, Research and Marketing are individually callable in the bounded local profile. Complete Workflow adoption in a commissioned external system remains unqualified.
- No live service is connected: no real GitHub read, error tracker or analytics source has been used.
- Power loss, other platforms and more than one machine are not tested.
- The sandbox is not a general sandbox for arbitrary code: there is no hard memory, disk or CPU-time stop.
- Records are attributed, not authenticated: anyone who can write the local stores could forge consistent records.
What comes next, in order
-
In development
-
Deliver one useful Slice for a commissioned real Product, with independent evidence.
In development
-
Qualify bounded, proposal-only model routes through your subscriptions.
In development
-
Plan from each product's saved values.
In development
-
Add the evidence collectors a real Slice needs, then bounded evolution.
Planned
-
Then discovery before building, and release and measurement after it.
In development
Progress log
Each recorded step, newest first, with the evidence behind it. The docs explain every module and its limits.
-
Bounded local development complete
Independent qualification passed for 36 installed module families, callable stages and 2 full 13-stage workflow rehearsals. The accepted source passed 3321 root and 27 Dashboard checks; merged main passed its own full CI. This proves the bounded local profile, with synthetic Products and controlled collaborators. All 42 epic gates, live providers, customer delivery and release remain open.
Evidence: Current status and next steps · Joint qualification (JSON) · Build review
-
Visual guides and the Plumb family published
The docs now explain module inputs, outputs and reuse boundaries with diagrams. The module portraits are live in the catalogue and on How it works. These publish the explanation and artwork; they do not establish customer use.
Evidence: Visual understanding · Plumb family
-
The adoption work accepted locally, family by family
The owner made the coordinating Claude Code session the acceptance reviewer. After the independent Claude Opus 5.5 reviews and an advisory Fable 5.1 review of each, it accepted the 24 module families of the merged adoption work locally, each with a new receipt that pins its files; earlier drafts and receipts are unchanged. This covers local infrastructure only: required CI, live use, release and customer value are not established, and every epic stays open.
Evidence: Build review: adoption-stack acceptance
-
The adoption work merged into the main branch
With the planned acceptance reviewer unavailable, the owner had Claude Opus 5.5 review the 31 stacked drafts in order, then authorised one merge that keeps every commit. Two drafts needed fixes first, and both were fixed before the merge; the rest could merge, most with follow-ups now tracked as issues. Merging did not accept them; their acceptance, with a receipt for each family, followed later that day.
Evidence: Handover: next steps (not published) · Implementation plan · Build review
-
Models named in priority order
The owner decided that the Claude command line stays unavailable for Factory's model roles, and that discovery's sorting and weekly synthesis should try Codex first. The model registry can now name up to four models in order, takes the first that nothing recorded refuses, and never lets a checker share a model with what it checks. The first independent review round failed on two findings, both fixed before the merge. Nothing is qualified, so nothing runs.
Evidence: Intelligence: priority lists · Intelligence guide · Build review
-
Dates, capacity and timing, checked without inventing certainty
Adoption epic E36 covers planning. Its first three evaluators passed independent review: one keeps each date as precise as its source and shows conflicts, one gives a capacity range only from comparable reviewed work, and one checks a proposed time against notice periods, embargoes and clock changes. None stores a plan or starts work.
Evidence: Temporal Planning guide · Capacity planning guide · Material timing guide · Build review
-
Service failures as Signals, and a design for repairing them
Adoption epics E37 and E38 cover taking in what a product's services report, and repairing it through verified changes. The intake passed independent review on test data. No live service is connected, and the Repair Workflow is designed, not built.
Evidence: Operational signals guide · Intake and Repair design
-
Factory's parts as packages, and a digest of what needs you
Drafts that package the foundational modules for use outside Factory, each proved by a separate consumer, alongside Branch health, the owner digest, the ask ledger and an eight-value version of each product's guidance. Each passed independent review on recorded or test data; none reads a live service.
Evidence: Adoption inventory · Owner digest guide · Branch health guide · Build review
-
A delivery plan for independent adoption
Eight adoption epics, E28 to E35, now track the work needed to use Modules and complete Workflows in another system, alongside the original 27 customer-value epics. The plan separates existing code refactors, new Workflows and Factory integration proof. Estimates are provisional; independently adoptable distributions and complete Workflows remain unproved.
Evidence: Independent adoption · Implementation guideline
-
The first parts of discovery: a schedule and a page reader
Lifecycle increments 1 and 2 of the proposed discovery design: a schedule that names each nightly and weekly pass once and holds its token allowance, and a reader that fetches only commissioned public pages, reads robots.txt first and keeps quotes that can be checked again. Each passed independent review, the schedule on the fourth round and the reader on the fifth. Both are tried only on test data, and nothing runs on a schedule.
Evidence: Lifecycle schedule guide · Discovery sources guide · Build review
-
The sandbox collector and a last check before a branch move
Step 1, increments 3 and 3b: the Repository collector runs a Candidate's frozen checks in the sandbox and stores what it saw, and Repository can run a last check just before it moves a branch. Both accepted on their second review round; neither is joined into delivery yet.
Evidence: Repository collector guide · Repository guide · Build review · Last-check receipt (JSON)
-
Repository Assurance and the Product gate
Step 1, increments 1 and 2: a judge for repository Candidates, and one durable record per product for grants, admissions and delivery records. Accepted on the sixth review round, after rounds 1 to 5 each found something to fix. Not yet joined into delivery.
Evidence: Repository Assurance guide · Product gate guide · Build review
-
Subscriptions only, and the full lifecycle
The owner directed that models run only through the owner's own subscriptions, never Platform API keys, and added discovery before building and release, measurement and clean-up after it. Recorded in the project intent; its first parts are now built (see above).
Evidence: Project intent (not published)
-
An explicit decision about models, per module
Every module now declares whether it uses a model, which one and why; most use none. Accepted on review; no model is qualified yet, so every model task refuses.
Evidence: Intelligence guide
-
Producers for Claude and OpenAI
One contract for asking a model, once, for bounded edit proposals, with adapters for the Claude Code command line and OpenAI's Responses API. Accepted on the ninth review round. Both live routes are unavailable.
Evidence: Producers guide · Claude receipt (JSON)
-
The local dashboard, Portfolio and project values
Six local products registered for read-only metadata; a dashboard of sourced progress where unknown is never zero; each product's core values kept as exact revisions; one list of what needs attention. Checked in a real browser at desktop and phone widths.
Evidence: Dashboard guide · Portfolio guide · Project guidance guide · Dashboard receipt (JSON) · Portfolio receipt (JSON) · Guidance receipt (JSON)
-
Sealed Candidates, a macOS sandbox and invoke-once integration
Exact, immutable Candidates in repositories that Factory owns; a sandbox that runs one bounded program and reports only what it saw; a branch move made at most once, reporting unknown when unsure.
Evidence: Repository guide · Confined executor guide · Sealing receipt (JSON) · Executor receipt (JSON) · Integration receipt (JSON)
-
Records, runs and recovery
The command journal, Evidence and Verdicts, Run and Attempt ownership, the fixture worker, per-Attempt fences and crash recovery within Budget: what every later module depends on.
Evidence: Command journal guide · Guarded verification guide · Baseline receipt (JSON) · Recovery receipt (JSON)
-
The design, its values and the original 27 customer-value epics
The initial design established one architecture at three reading depths, one domain language, core values and 27 customer-value epics, each with measurable success criteria.
Evidence: Design · Values · Implementation guideline
