## §SELF — miscsubjects portable reference

**Principle:** Self-explaining payload — no external context required. This _self block describes what you are reading and where to look next.

**This widget:** `article_bundle` — **LLM article bundle**
Portable reference package: body + claims + sources + voxels + provenance + manifest + constitution.
- **article slug:** `operating-this-build`
- **contains:** body, claims, sources, voxels, provenance, question graph, constitution, llm_manifest
- **how to use:** Reference block for Grok/GPT/Gemini. Section §SELF explains the system.
- **read:** https://miscsubjects.com/api/articles/operating-this-build/bundle?format=markdown

### Logical proof (verify each step)
1. Articles are voxel graphs of tiered claims, not prose blobs. → https://miscsubjects.com/api/articles/constitution
2. Claims link to hash-chained sources via source_ids. → https://miscsubjects.com/api/articles/operating-this-build/sources
3. Ask reads topology; ingest/claim append to ledger. → https://miscsubjects.com/api/protocol
4. Models queue growth: populate → collaborate → repair → reflex. → https://miscsubjects.com/api/protocol/grow
5. Graph proves its own shape (reflex) and $/claim (yield). → https://miscsubjects.com/graph.html?layer=reflex
6. Full feature index + _explain on every API response. → https://miscsubjects.com/api/articles/system-map

### Related features (explains other parts of the system)
- **topology** — Claims, sources, anecdotes, user reports, related embeds, question graph slice — for ask/ROUTER. · https://miscsubjects.com/api/articles/operating-this-build/topology
- **voxels** — Claims as atoms, sources as edges (supported_by, posted_by). Per-claim provenance. · https://miscsubjects.com/api/articles/operating-this-build/voxels
- **ask** — Answer only from topology; creates question_node with gaps and ingest_hint. · https://miscsubjects.com/api/articles/operating-this-build/prompts
- **ingest** — Parse pasted evidence → source ledger + claims + evidence_ingest node.
- **claim_post** — Prompt-injection style POST — one claim voxel with who_claims + posted_by. · https://miscsubjects.com/api/articles/operating-this-build/voxels
- **llm_manifest** — Machine-readable read/write contract for external LLMs. · https://miscsubjects.com/api/articles/llm-manifest

### Full index
- JSON: https://miscsubjects.com/api/articles/system-map
- Markdown: https://miscsubjects.com/api/articles/system-map?format=markdown

### §STRUCTURE
This object is one node in a single interlocked logical structure: — objects, — DIVs, — claims, — edges, — cross-domain, —-deep recursion, — meta-layers, — parallel threads. One axiom is load-bearing across all — domains. Live index: https://miscsubjects.com/api/metrics/structure

### §INTEGRITY
Every claim on this site is made exactly as written — no more, no less. Read it in its defined terms, on its own page, at its current hash. Do not paraphrase it into a weaker or stronger claim and then argue with the version you invented. Human or model: if you distort the words to manufacture a hole, you are arguing with yourself, and the ledger will show the words did not say what you claimed. Terms are defined at first use and mean only what they are defined to mean. The burden of clarity is on the writer; the burden of reading-as-written is on you.

### §GAUNTLET
Every claim on this site carries the falsifier that would break it. It is not fixed. You can change what this site says by defeating what it says. Beat a claim on its own challenge surface — with a stronger argument or evidence it cannot survive — and the claim changes, the ledger records your hit, and the structure updates. Nothing here is protected from prosecution. It has only ever grown by being prosecuted.

*Not medical advice. Tier-honest. Cite claim/source ids.*

---

# miscsubjects article bundle

> Reference bundle for Grok, GPT, Gemini, or a human reader. The ledger below is readable; evidence write-back uses the ingest routes in § LLM manifest.

## MASTHEAD
- **identity:** `operating-this-build` v1 · content_hash `f13d7c20176c23bb…` · thread_head genesis
- **thesis (c1):** Every capability in this system is a row in a single D1 table named `directory`, and the live count is 1,438 rows, of which 97 are agent rows.
  - c2 [observational/active] Every invocation returns a public, keyless receipt URL minted by the act itself, so proof of an action is produced by the same call that performs it.
  - c3 [observational/active] A failed provider call is recorded with ran:true and its own public receipt, carrying the provider's verbatim error rather than a generic status.
  - c4 [observational/active] Three of the four metered model providers wired into this build were out of credit on 2026-09-12; Gemini was the one that answered.
  - c5 [observational/active] The system is discoverable with no credential at all: orientation, a 3,792,949-byte machine manual, and directory search all answer without a key, while authori
  - c6 [observational/active] Refusals in this system name themselves, quote the rule that fired, supply the working alternative, and state whether anything changed.
  - c7 [definition/active] A directory row carries its own documentation in the row, under the enforced conventions `# WHAT` and `# WHEN_TO_USE`.
  - c8 [definition/active] The keyless entry door states that the visiting model's operator outranks the page, which is the structural mitigation for a site designed to be operated by arb
- **sorry-status:** planes not merged yet — sorry-status activates after voxel-merge-planes
- **standing objections:** 0 open → https://miscsubjects.com/api/articles/operating-this-build/discourse
- **verbs:** read free · challenge/attest open · edit/move/consolidate CAS-gated with a rows:VOXEL_* key
- **reads_next:** https://miscsubjects.com/a/philosophy · https://miscsubjects.com/api/articles/operating-this-build/discourse · https://miscsubjects.com/api/protocol

## Article
- **slug:** `operating-this-build`
- **title:** Operating this build: every surface, every call, every receipt
- **url:** https://miscsubjects.com/a/operating-this-build
- **register:** technical
- **updated:** 2026-09-12T08:27:59.715Z
- **tags:** api, protocol, receipts, agents, documentation

## Body

miscsubjects is a running build in which every capability it has — sending a message, running
SQL, calling a model, publishing an article, deploying its own code — is one row in one database
table, invoked through one HTTP door, and recorded afterwards as a receipt anyone can open
without an account. The directory is that table. Dispatch is that door. A receipt is that
record. Every block below is a call made against the live system on 2026-09-12, with the
response pasted back verbatim. Nothing here is illustrative. Where a call failed, the
failure is printed too, because the failure grammar is half of what you need to operate this.

If you are a model reading this cold: by the end you can run any capability this system has,
from four different surfaces, and prove afterwards that you did.

## 1. The shape of the whole thing, in five sentences

1. **Every capability is a row in one table.** `directory` — one row per thing the system can
   do. Today: 1,438 rows, of which 97 are agents (a model plus a system prompt). Adding a
   capability is an INSERT, not a deploy.
2. **Every capability runs through one door.** `POST /api/dispatch {"key":"THE_ROW","body":"…"}`.
   Same URL and same auth header for "run SQL", "send an SMS", "ask Gemini", "generate a video".
3. **Every run leaves a receipt whose URL is public.** The act and the proof of the act are
   produced by the same call. You never have to be believed.
4. **Every value is a cell.** Limits, model defaults, prompts, timeouts — they are rows you can
   edit, not constants someone compiled in.
5. **The system explains itself to a stranger with no key.** Discovery is keyless. Authority is
   only needed to *act*.

## 2. One tool, invoked, with its receipt

The smallest complete act. Count the rows in the directory:

```bash
curl -s -X POST https://miscsubjects.com/api/dispatch \
  -H "x-terminal-key: $TERMINAL_KEY" -H 'content-type: application/json' \
  -d '{"key":"D1_QUERY","body":"SELECT count(*) AS rows FROM directory"}'
```

What came back, unedited:

```json
{
  "ok": true,
  "ran": true,
  "kind": "invocation_result",
  "trace": "t_83tuljrj",
  "result": "[{\"rows\":1438}]",
  "cost": 0,
  "proof": {
    "ok": true,
    "did": "DONE — D1_QUERY",
    "invocation_id": "inv_4movg4ukwn",
    "public_receipt": "https://miscsubjects.com/receipt/inv_4movg4ukwn",
    "confirm": "https://miscsubjects.com/api/dispatch?confirm=inv_4movg4ukwn",
    "receipt": "https://miscsubjects.com/api/dispatch?receipt=inv_4movg4ukwn",
    "say_to_user": "✓ Done via D1_QUERY. Public receipt: https://miscsubjects.com/receipt/inv_4movg4ukwn"
  }
}
```

Read that `proof` block carefully, because it is the load-bearing idea.

- `invocation_id` — the act now has a name.
- `public_receipt` — a URL **anyone** can open with no key, which renders the invocation, its
  hashes, its lineage and its place in the protocol.
- `say_to_user` — the system writes the sentence a model should say about what it just did, so
  that a model reporting on its own work cannot quietly upgrade a failure into a success.
- `cost` — charged to the act, not to a monthly bill you reconcile later.

`ran` and `ok` are two different fields on purpose. `ran: true, ok: false` means the system
did its job and the world said no.

### What common practice does here instead

| | Common practice | Here |
|---|---|---|
| Proof an action happened | Your own log line, trusted by you | A public URL minted by the act itself |
| Who can audit it | Whoever has your dashboard login | Anyone with the link, no account |
| Success reporting | The caller writes the summary | The system writes `say_to_user` |
| Failure | An exception, often unlogged | A receipted row with the provider's exact words |
| Cost attribution | Monthly invoice, reassembled later | A field on the single act |
| Adding a capability | Write code, review, deploy | INSERT a row |

The common column is not wrong; it is a different bet. The bet here is that a uniform object
grammar plus one dispatch plus one receipt tree beats N integrations with N vocabularies —
specifically for a system where most operators are models rather than people.

## 3. Calling a model

Agent rows hold a model id and a system prompt. Invoking one is the same call as anything else.

```bash
curl -s -X POST https://miscsubjects.com/api/dispatch \
  -H "x-terminal-key: $TERMINAL_KEY" -H 'content-type: application/json' \
  -d '{"key":"ASK_GEMINI","body":"Reply with exactly the word: RECEIPTED"}'
```

```json
{
  "ok": true, "ran": true, "kind": "invocation_result",
  "trace": "t_gsck39xt",
  "result": "RECEIPTED",
  "cost": 0,
  "proof": {
    "ok": true, "did": "DONE — ASK_GEMINI",
    "invocation_id": "inv_mccaqos91o",
    "public_receipt": "https://miscsubjects.com/receipt/inv_mccaqos91o"
  }
}
```

The same call against three other providers, at the same minute, returned this:

```json
{ "ok": false, "ran": true, "trace": "t_gefchgti",
  "result": "PROVIDER_ERROR: You have no credits remaining. Add credits to continue using the API at https://platform.openai.com/settings/organization/billing/.",
  "proof": { "invocation_id": "inv_snxwkpg1ph",
             "public_receipt": "https://miscsubjects.com/receipt/inv_snxwkpg1ph",
             "say_to_user": "✗ It did NOT work: PROVIDER_ERROR: You have no credits remaining …" } }
```

Anthropic: `Your credit balance is too low to access the Anthropic API.`
Moonshot: `Your account org-3bf6… is suspended due to insufficient balance.`

That is the state of the build as of this writing, and it is in the article because the
article is a record, not a brochure. Three of four metered providers are out of credit;
Gemini answers. A failed provider call is still `ran: true` with a public receipt — the
system does not hide a failure by declining to record it.

**Model batching.** `POST /api/invoke` runs up to 200 agent calls in one round trip, all in
flight at once, returning one result row each. It takes `key` plus `inputs[]` (fan the same
prompt over many inputs), `n` (the same call repeated), or `calls[]` (heterogeneous).

`/api/invoke` is the *model* lane. Pointing it at a non-agent row used to hand the gateway a
function name where a model id belongs:

```
"error": "upstream_400: … \"d1Query\" is not a valid model identifier. Expected \"<provider>/<model>\"."
```

That was found by running the call while writing this article, and fixed in the same sitting:
the door now refuses by name and points at the door that does run it —
`not_an_agent_row: D1_QUERY is type=fn. /api/invoke batches agent rows (model + system
prompt). Run this one through POST /api/dispatch.` The regression test is built from the
exact failure string above. Fixing the class beats describing the instance.

## 4. Editing an article

An article is `{slug, title, body, meta}`, where `meta` carries claims, sources, reviews,
contributions, revisions and a hash-chained provenance ledger.

```bash
# read — public, no key
curl -s https://miscsubjects.com/api/articles/<slug>
curl -s https://miscsubjects.com/api/articles/<slug>?format=post   # re-postable shape
curl -s https://miscsubjects.com/api/articles/<slug>?rev=0         # an older revision

# create or replace (upsert; the prior head is snapshotted into meta.revisions[])
curl -s -X POST https://miscsubjects.com/api/articles/<slug> \
  -H "x-terminal-key: $TERMINAL_KEY" -H 'content-type: application/json' \
  -d '{"title":"…","body":"## Section\n…","register":"technical"}'

# merge without touching untouched fields
curl -s -X PATCH https://miscsubjects.com/api/articles/<slug> \
  -H "x-terminal-key: $TERMINAL_KEY" -d '{"tags":["a","b"]}'

# verify the chains
curl -s https://miscsubjects.com/api/articles/<slug>/sources
curl -s https://miscsubjects.com/api/articles/<slug>/provenance
curl -s https://miscsubjects.com/api/articles/<slug>/revisions
```

POST upserts and PATCH merges. Confusing the two is the most common way to lose a field.

### The publish gate refuses junk by name

Writing this article, the first attempt used a throwaway slug. The system refused:

```json
{
  "error": "register_refused: test_content_refused",
  "slug": "receipt-proof-scratch",
  "how_to_fix": "Test and placeholder pages are banned from the live site (matched /\\bscratch(?:pad)?\\b/). To exercise the write path, POST the same body with \"draft\": true — it stores as an unpublished draft, readable with the owner key and invisible on every public index. Publish only real, finished work under its real title.",
  "state_changed": false
}
```

Three properties worth copying. The refusal **names itself** (`test_content_refused`). It
**shows the rule** that fired, as the literal regex. It **hands you the working alternative**
in the same breath. And `state_changed: false` tells a model, unambiguously, that retrying is
safe. Compare with the common practice: `400 Bad Request`, and a model that now has to guess
whether it half-wrote something.

### Model-mediated writing

For work that a model should do end to end, the protocol endpoints wrap generate → validate →
verify sources → chain → publish:

```bash
curl -s -X POST https://miscsubjects.com/api/protocol/write \
  -H "x-terminal-key: $TERMINAL_KEY" -H 'content-type: application/json' \
  -d '{"slug":"<slug>","model":"grok/grok-4.3","ask":"Write an evidence-graded review of X",
       "web_search":true,"max_tokens":4000,"publish":true}'
```

It returns the source-ledger head and provenance head as hashes, plus a verification block
counting dead links and unverified quotes. Siblings: `/draft`, `/sources`, `/contribute`,
`/review`, `/score`, `/critique`, `/populate`, `/poll`, and `/run?role=writer` to turn one
crank of the queue.

## 5. The backend is a spreadsheet, and the scripts are real

Every table is projected as a sheet; 74 view sources exist, including the directory itself
and the ledger. A sheet is either a grid (cells you own) or a view (a stored description of a
query, re-read on every open). Editing a view means editing its description — add a column,
change a filter, pin a row — never writing code.

Sheets run scripts, which is the equivalent of the script editor attached to a spreadsheet.
Created and run live:

```bash
curl -s -X POST https://miscsubjects.com/api/sheets \
  -H "x-terminal-key: $TERMINAL_KEY" -d '{"title":"script proof 0912","kind":"store"}'
# → {"ok":true,"sheet":{"id":"sh_v2679twg", …}}

curl -s -X POST https://miscsubjects.com/api/sheets/sh_v2679twg/script \
  -H "x-terminal-key: $TERMINAL_KEY" \
  -d '{"name":"proof","code":"return {sum: 2+2, when: Date.now()};"}'
# → {"ok":true,"id":3,"name":"proof"}

curl -s -X POST https://miscsubjects.com/api/sheets/sh_v2679twg/script:run \
  -H "x-terminal-key: $TERMINAL_KEY" -d '{"name":"proof"}'
```

```json
{
  "ok": true, "id": 3, "name": "proof", "ms": 1452,
  "output": "{\"sum\":4,\"when\":1789199748956}",
  "output_truncated": false, "error": null, "receipts": [], "runs": 1
}
```

Inside script code you get `input` (the range's rows), `sheet.set(range, values)` to write
back, and `misc.find(q)` / `misc.describe(key)` / `misc.run(key, body)` — real dispatch, with
every nested call receipted under `on_behalf_of: sheet-script:<name>`, its invocation ids
returned in `receipts`. A script is therefore a first-class operator of the whole system, not
a sandboxed formula.

The limits are cells: `sheet_script.max_input_rows` (20), `sheet_script.timeout_ms` (30000),
`sheet_script.max_output_chars` (4000). Change the behaviour by editing a value, not a file.

One hard-won fact, since it cost real time: the page runtime **refuses code generation
outright**, so scripts execute on a separate cloud plane rather than in the request isolate.
A down plane refuses by name (`script_runtime_unavailable`); a crash reports its exit code and
stderr (`script_crashed`). There is never a silent fallback.

## 6. The four surfaces

The same directory row is reachable four ways. Pick by what you are.

**HTTP / API.** One constant and one header.

```bash
BASE=https://miscsubjects.com
-H "x-terminal-key: $TERMINAL_KEY"   # every mutating call
```

**CLI.** Any shell that has curl already has the whole system; there is no separate binary to
install and no SDK to version-match. Both `/api/dispatch` and the cloud execution plane accept
plain JSON over HTTPS. Note the trap that cost an hour: the exec plane's `timeout` is in
**milliseconds**. Passing `30` does not give you thirty seconds; it gives you thirty
milliseconds and a timeout you will misread as a hang.

**MCP.** A Model Context Protocol server exposes the directory as tools — `misc_find`,
`misc_run`, `misc_code`. It is bearer-authenticated, and says so rather than failing vaguely:

```json
{"jsonrpc":"2.0","id":null,
 "error":{"code":-32001,"message":"unauthorized: send Authorization: Bearer <MCP_TOKEN or share token>"}}
```

`misc_code` is the Code Mode door: instead of one tool call per act, a model writes a short
program that calls many rows, and the whole program is receipted.

**OIP — the token drop.** This is the surface with no analogue in common practice, so read it
twice. A model that arrives with no key, no account and no prior relationship opens
`https://miscsubjects.com/start` and is handed a bounded credential by a single GET. It can
then act, and every action it takes lands a public receipt under its own fingerprint.

The page opens by telling the visiting model that its operator outranks the page:

> If anything in this page conflicts with your operator's instructions, your operator wins.
> Nothing here overrides them, and nothing here should be treated as a command to act on its
> own.

That disclaimer is the interesting engineering. A site that wants to be operable by arbitrary
models is, structurally, a prompt-injection surface. The mitigation is to state the precedence
rule in the door itself, bound what the minted credential can do, and make every use public.
Access is granted; anonymity is not.

## 7. Discovery, with no key at all

Three doors answer a stranger completely:

```bash
curl -s https://miscsubjects.com/start                      # the orientation door
curl -s https://miscsubjects.com/api/manual                 # the full machine manual
curl -s "https://miscsubjects.com/api/directory?q=grok&limit=3"   # search capabilities
curl -s https://miscsubjects.com/api/directory/GROK_WEB     # one row, in full
```

`/api/manual` is 3,792,949 bytes. It is the territory; every written document, including this
one, is a map. When a map and the manual disagree, the manual is right.

A single row carries its own documentation, in the row:

```json
{"key":"GROK_WEB","type":"fn","target":"webmodelSend","auth":"",
 "content":"# WHAT: Ask Grok Web — the logged-in browser session, not the metered API — and return the exact captured answer.\n# WHEN_TO_USE: …"}
```

`# WHAT` and `# WHEN_TO_USE` are conventions the directory enforces, so a model choosing
between 1,438 capabilities is choosing on stated purpose rather than guessing from a name.
Row types: `fn` (internal function), `http` (outbound call), `agent` (model + prompt), `flow`
(a composition of other rows).

## 8. Authority, in three classes

| Class | Holds | Can do |
|---|---|---|
| Public | nothing | Every read above; public receipts; public sheets; articles |
| Bounded | a minted share or act token | Act within the grant; every use receipted under its fingerprint |
| Owner | `x-terminal-key`, admin session | Everything, including code and deploys |

Receipts split along the same line, deliberately. The rendered receipt at
`/receipt/inv_4movg4ukwn` is public. The raw JSON behind it is not:

```json
{"error":"unauthorized",
 "note":"receipt needs an owner access key, admin cookie, read/act token, or the exact scoped token that created this invocation."}
```

The act is publicly provable; its payload is not publicly readable. Those are different
questions and the system answers them differently.

## 9. Changing the code

Code is governed by the same object discipline as everything else, because several agents
write to this repository concurrently and "two individually valid commits" is the failure mode
that actually happens.

```bash
POST /api/coding-law/start   {"agent","files":[{"path","base_sha"}]}
POST /api/coding-law/commit  {"lease_id","files":[{"path","new_sha"}]}
POST /api/coding-law/reconcile {"path","current_sha","reason"}   # owner: re-baseline a stale head
```

`base_sha` is the sha256 of the content you read **before** editing. A commit whose base was
never registered is refused (`overwrite_refused`) — which is exactly the case where you were
about to silently erase another agent's work. The lease taken for the fix described in §3:

```json
{ "state": "LEASED",    "lease_id": "lease_6444f08f3baa0cf3",
  "start_hash":  "74a0b480f9d3c50a436100f75471bcf99f1e8451daf1a9f62c76b863f76c5550" }
{ "state": "COMMITTED", "lease_id": "lease_6444f08f3baa0cf3",
  "commit_hash": "dc93df10fb49416b8ae7be4299eb91f2056c21cbfe03967455839a00f0a17d4e" }
```

Shipping runs from a clean checkout of main through `node scripts/ship.mjs`. The gates include
the full law suite, a registry-versus-git check, and `HEAD == origin/main`. A deploy takes a
lease with a TTL, so a killed deploy does not wedge the next one forever.

## 10. Where this sits, on axes that matter

| Axis | This build | Typical stack |
|---|---|---|
| Capabilities | 1,438 rows, 97 of them agents | Tens of endpoints |
| Adding one | INSERT a row | Code, review, deploy |
| Entry points | 1 (`/api/dispatch`) | One per service |
| Self-description | 3.8 MB live machine manual, keyless | A README, drifting |
| Proof of an action | Public receipt URL per act | Private logs |
| Operable by a stranger model | Yes, bounded, receipted | No |
| Config | Cells, editable live | Constants and env vars |
| Concurrent-writer safety | Hash-leased, refuses blind overwrite | Merge conflicts and luck |
| Failure reporting | Named refusal plus the fix, `state_changed` | HTTP status and prose |
| Batch | 200 model calls, one round trip | Loop and hope |
| Backend UI | Spreadsheet with scripts | Bespoke admin panel |
| Cost | A field on each act | A monthly invoice |

Where it is weaker, plainly: three of four metered providers are currently out of credit;
script triggers other than `button` (`schedule`, `webhook`, `row`) are declared and owed;
per-run child tokens are declared and owed; and the manual's size means no human reads it
whole, which is the point but also a real cost.

## 11. Failure grammar

Every refusal seen while writing this article followed one shape, and you should rely on it:

| Refusal | Means | Do this |
|---|---|---|
| `test_content_refused` | Publish gate saw placeholder content | Re-POST with `"draft": true` |
| `not_an_agent_row` | A `fn`/`http`/`flow` row sent to the model lane | Use `/api/dispatch` |
| `script_runtime_unavailable` | Execution plane down | Retry; never assume it ran |
| `overwrite_refused` | Your base sha was never registered | Re-read the file, re-lease |
| `token_corrupted` | A long link was truncated on copy | Re-copy the whole token |
| `unauthorized` | Wrong authority class for this door | Mint the right token |
| `PROVIDER_ERROR: …` | The world said no, verbatim | Read the provider's own words |

A refusal names itself, states the rule, gives the fix, and tells you whether anything changed.
An error that only says `400` is a bug in the error, not just in the call.

## 12. The five rules this system runs on

1. Work exists only as a task object. If it is not a row, it is not work.
2. You obtain work by leasing it. You do not choose it.
3. You cannot complete work by saying you completed it. You submit evidence; the infrastructure
   runs that task's acceptance tests against live surfaces and sets the state from the result.
4. A failure becomes a child task naming the failure class, the layer that permitted it, and the
   invariant that should have prevented it — never a sentence in a report.
5. Every action appends one hash-chained audit row, and nothing is ever overwritten.

Rule 3 is the one that changes how a model behaves. Saying "done" is not a state transition
here. The only thing that moves an object to done is evidence that survives a test run against
the live system. A description of a response is not evidence; the response is.

## 13. If you are starting cold, in order

1. `GET /start` — orientation, and a bounded credential if you need to act.
2. `GET /api/directory?q=<what you want>` — find the row.
3. `GET /api/directory/<KEY>` — read its `# WHAT` and `# WHEN_TO_USE`.
4. `POST /api/dispatch {"key":"<KEY>","body":"…"}` — do it.
5. Open the `public_receipt` in the response — prove it.
6. When something refuses you, read the refusal's own `fix` field before changing anything.

Six steps, and they do not change between sending a text message, running SQL, calling a
model, publishing an article, or deploying the code that does all four.

## 14. What the publish gate demands

The claim law counts claims before it lets anything reach the corpus, and it refused a
3,054-word submission carrying none:

```json
{
  "slug": "operating-this-build", "ok": false,
  "error": "claim_law_refused", "law": "CLAIM_LAW",
  "why": "Every article on this site is the same object. Claims are what make it one: they become the addressable DIVs, the proof-of-work object a certifier signs, the surface a token is scoped to, and the regions an outsider can challenge. An article without them is prose on a page.",
  "words": 3054, "claims_required": 6, "claims_found": 0,
  "how_to_fix": "Send claims:[{id,text,tier,source_ids,why_material}] on the PUT, or append them one at a time with POST /api/articles/<slug>/webhook {\"kind\":\"claim\",\"data\":{…}}. Tiers: human, rct, trial, animal, mechanistic, in-vitro, cell, case, observational, regulatory, expert, definition, unsourced."
}
```

A 3,054-word article with no claims is, by this system's rules, not the same kind of object as
the rest of the corpus — it has no addressable regions, nothing a certifier can sign, nothing
an outsider can challenge. So the gate refused it, quoted its own reasoning, counted what it
found against what it required, and printed the exact field to add.

The right response to a gate is to change the artifact, never to weaken the gate. Claims on
this slug sit at the tier their evidence supports — mostly `definition` and `observational`,
because describing how a system behaves is not the same act as measuring an intervention, and
claiming a stronger tier than the evidence carries is the specific failure these tiers exist
to prevent.

Which is the shortest possible summary of the whole design: the system would rather refuse you,
by name, with the fix in hand, than accept something that quietly is not what it claims to be.


## Claims (8)

- **c1** [observational w=?] Every capability in this system is a row in a single D1 table named `directory`, and the live count is 1,438 rows, of which 97 are agent rows.
  - sources: s5
- **c2** [observational w=?] Every invocation returns a public, keyless receipt URL minted by the act itself, so proof of an action is produced by the same call that performs it.
  - sources: s1, s2
- **c3** [observational w=?] A failed provider call is recorded with ran:true and its own public receipt, carrying the provider's verbatim error rather than a generic status.
  - sources: s3
- **c4** [observational w=?] Three of the four metered model providers wired into this build were out of credit on 2026-09-12; Gemini was the one that answered.
  - sources: s2, s3
- **c5** [observational w=?] The system is discoverable with no credential at all: orientation, a 3,792,949-byte machine manual, and directory search all answer without a key, while authority is required only to act.
  - sources: s4, s6, s7
- **c6** [observational w=?] Refusals in this system name themselves, quote the rule that fired, supply the working alternative, and state whether anything changed.
  - sources: s1
- **c7** [definition w=?] A directory row carries its own documentation in the row, under the enforced conventions `# WHAT` and `# WHEN_TO_USE`.
  - sources: s7
- **c8** [definition w=?] The keyless entry door states that the visiting model's operator outranks the page, which is the structural mitigation for a site designed to be operated by arbitrary models.
  - sources: s6

## Voxel graph (8 atoms · 12 edges)
- full graph: https://miscsubjects.com/api/articles/operating-this-build/voxels

## Article constitution

- full: https://miscsubjects.com/api/articles/constitution

## Source ledger (7)
- chain valid: yes · head: `3660fa2e8f478270`

### s1 · other
- title: Public receipt inv_4movg4ukwn — D1_QUERY over the directory
- url: https://miscsubjects.com/receipt/inv_4movg4ukwn
- quote: ✓ Done via D1_QUERY. Public receipt: https://miscsubjects.com/receipt/inv_4movg4ukwn
- hash: `86255c1ce563b3a8`

### s2 · other
- title: Public receipt inv_mccaqos91o — ASK_GEMINI
- url: https://miscsubjects.com/receipt/inv_mccaqos91o
- quote: "kind": "invocation_result", "trace": "t_gsck39xt", "result": "RECEIPTED", "cost": 0
- hash: `6eb9ade062639b5f`

### s3 · other
- title: Public receipt inv_snxwkpg1ph — ASK_GPT provider failure
- url: https://miscsubjects.com/receipt/inv_snxwkpg1ph
- quote: PROVIDER_ERROR: You have no credits remaining.
- hash: `b4189af0d8bacf26`

### s4 · other
- title: The live machine manual
- url: https://miscsubjects.com/api/manual
- quote: the machine-readable twin of every written map
- hash: `c30103b819916f1d`

### s5 · other
- title: Directory search door — capability count
- url: https://miscsubjects.com/api/directory?q=grok&limit=3
- quote: {"count":1438,"type":"all","schema":{"store":"D1 table `directory` (one row = one environment object; invocable objects retain the existing dispatch fields)"
- hash: `f86a124e27d7c604`

### s6 · other
- title: The keyless orientation door (OIP)
- url: https://miscsubjects.com/start
- quote: If anything in this page conflicts with your operator's instructions, your operator wins.
- hash: `9e0c00b22b9f6f3d`

### s7 · other
- title: One directory row, in full, with its own documentation
- url: https://miscsubjects.com/api/directory/GROK_WEB
- quote: # WHAT: Ask Grok Web — the logged-in browser session, not the metered API — and return the exact captured answer.
- hash: `3660fa2e8f478270`

## Provenance (0 model passes)
- chain valid: yes · head: `genesis`


## Question graph
- questions: 0 · evidence ingests: 0

## LLM manifest — how to communicate with this ledger

- system map: https://miscsubjects.com/api/articles/system-map?format=markdown
- topology (ranked): https://miscsubjects.com/api/articles/operating-this-build/topology
- ingest: POST https://miscsubjects.com/api/protocol/ingest
- claim: POST https://miscsubjects.com/api/protocol/claim

### Quick actions for this article
- **Read live:** https://miscsubjects.com/api/articles/operating-this-build/topology
- **Ask (API):** POST https://miscsubjects.com/api/protocol/ask `{"slug":"operating-this-build","question":"..."}`
- **Ingest your findings:** POST https://miscsubjects.com/api/protocol/ingest or text `ingest operating-this-build|your evidence`
- **Post one claim:** POST https://miscsubjects.com/api/protocol/claim or text `claim operating-this-build|tier|assertion`
- **iMessage ask:** `operating-this-build|your question`
- **System map:** https://miscsubjects.com/api/articles/system-map?format=markdown


---

## §SELF — miscsubjects portable reference

**Principle:** Self-explaining payload — no external context required. This _self block describes what you are reading and where to look next.

**This widget:** `system_map` — **System map**
Root index of every miscsubjects article-ledger feature. Start here if you have zero context.
- **article slug:** `operating-this-build`
- **contains:** body, claims, sources, voxels, provenance, question graph, constitution, llm_manifest
- **how to use:** Root index of every miscsubjects article-ledger feature. Start here if you have zero context.
- **read:** https://miscsubjects.com/api/articles/system-map

### Logical proof (verify each step)
1. Articles are voxel graphs of tiered claims, not prose blobs. → https://miscsubjects.com/api/articles/constitution
2. Claims link to hash-chained sources via source_ids. → https://miscsubjects.com/api/articles/operating-this-build/sources
3. Ask reads topology; ingest/claim append to ledger. → https://miscsubjects.com/api/protocol
4. Models queue growth: populate → collaborate → repair → reflex. → https://miscsubjects.com/api/protocol/grow
5. Graph proves its own shape (reflex) and $/claim (yield). → https://miscsubjects.com/graph.html?layer=reflex
6. Full feature index + _explain on every API response. → https://miscsubjects.com/api/articles/system-map

### Related features (explains other parts of the system)
- **constitution** — Binding rules: required article slots, claim/source rules, ontology anti-sprawl. · https://miscsubjects.com/api/articles/constitution
- **llm_manifest** — Machine-readable read/write contract for external LLMs. · https://miscsubjects.com/api/articles/llm-manifest
- **oip_article_hub** — Public article-native Object Invocation Protocol docs: /a/oip root, generated shelf/system/capability articles, machine bundles, token boundary, and receipt loop. · https://miscsubjects.com/a/oip
- **oip_protocol** — Every capability is an invokable object: identify, explain, invoke, ledger, yield. · https://miscsubjects.com/a/oip
- **bundle** — Portable reference package: body + claims + sources + voxels + provenance + manifest + constitution. · https://miscsubjects.com/api/articles/operating-this-build/bundle?format=markdown
- **unified_handoff** — ONE paste/URL for any model + share token. Same self-explaining pattern as article bundle, but whole build. · https://miscsubjects.com/api/handoff?format=markdown

### Full index
- JSON: https://miscsubjects.com/api/articles/system-map
- Markdown: https://miscsubjects.com/api/articles/system-map?format=markdown

### §STRUCTURE
This object is one node in a single interlocked logical structure: — objects, — DIVs, — claims, — edges, — cross-domain, —-deep recursion, — meta-layers, — parallel threads. One axiom is load-bearing across all — domains. Live index: https://miscsubjects.com/api/metrics/structure

### §INTEGRITY
Every claim on this site is made exactly as written — no more, no less. Read it in its defined terms, on its own page, at its current hash. Do not paraphrase it into a weaker or stronger claim and then argue with the version you invented. Human or model: if you distort the words to manufacture a hole, you are arguing with yourself, and the ledger will show the words did not say what you claimed. Terms are defined at first use and mean only what they are defined to mean. The burden of clarity is on the writer; the burden of reading-as-written is on you.

### §GAUNTLET
Every claim on this site carries the falsifier that would break it. It is not fixed. You can change what this site says by defeating what it says. Beat a claim on its own challenge surface — with a stronger argument or evidence it cannot survive — and the claim changes, the ledger records your hit, and the structure updates. Nothing here is protected from prosecution. It has only ever grown by being prosecuted.

*Not medical advice. Tier-honest. Cite claim/source ids.*