autoprompt

# Autoprompt

Autoprompt is a useful-first orchestration loop. It stores the mission once, produces one independently approved executable roadmap, dispatches implementation-ready lanes directly, and proves the delivered behavior with real tests and fresh review.

## 1. Start contract

The invocation authorizes the mission. Do not ask the user to restate, narrow, or approve it.

Loading the skill or invoking it without a mission never starts or resumes a run, regardless of leftover artifacts. A bare invocation performs only the section-10 frontier check, reports the result, and stops.

Before spawning, resolve only undefined operator knobs:

- **Concurrency:** `tokensaver` (default, up to six live), `wide`/`billionaire` (all ready disjoint work up to the global ceiling), or `custom max_subs=N`.
- **Agent selection:** `off`/inherit. VS Code subagents inherit the runtime model, so effort capability is exactly `inherited-only`; never claim a selectable effort.

In an attended session, ask all undefined knobs in one question before repository/tool work. In an unattended supervisor run, do not ask: default to `tokensaver` and `agents=off` and record the assumptions. A permission-bypass flag is not unattendedness.

After the chooser, dispatch the L1 scope coordinator `ap-scope-coordinator`; it dispatches the useful-first roadmap author. There is no separate intake round trip and no mandatory preflight agent.

An invoked mission always enters orchestration: the bounded scope topology is the minimum topology, and the main agent never silently self-triages an invoked mission into direct execution. Skipping or narrowing dispatch is an explicit decision recorded in `GATELOG.md` with its rationale, never a silent one.

## 2. Capability fast path

A supervisor may provide a versioned capability attestation bound to provider/runtime, CLI version, permission profile, agent selector, agent-definition hash, casting hash, effort status/source, and exact RUN/READ/WRITE success. Use it only when every binding matches the live launch. Missing, malformed, stale, unknown, or contradictory values are safe misses.

Without a trusted attestation, the first useful roadmap author proves RUN, READ, and WRITE against a disposable scratch path before repository inspection and then immediately continues. Any failure hard-stops before implementation. The preflight persona is diagnostic/recovery only.

## 3. Adaptive scope topology

Scope produces one canonical `ROADMAP.md`.

- **bounded:** roadmap author, then independent reviewer and blind fresh verifier concurrently - **3 agents, 2 rounds**, target under one minute;
- **multi-surface:** **exactly 5 agents, 3 rounds**, target under five minutes; retain the complete author roadmap and evidence, add exactly two complementary scouts, then concurrent reviewer plus fresh verifier without a redundant synthesis dispatch;
- **unusually-large:** may exceed the 6-agent ordinary budget only with a concrete recorded escalation reason.

External research runs only when current external facts are required. On rejection, retain accepted evidence and repair only named items. Empty roadmaps, invalid DAGs, overlapping ownership, missing frameworks/tests, and failed capability are hard failures.

## 4. Executable roadmap

`ROADMAP.md` is the new run's sole scope/decomposition/plan source. It includes mission pointer/hash and nonce; scope profile/escalation; repository intelligence; framework/tool decisions; stable item ids; category/tag/tier/framework; owned boundaries; dependencies and launch groups; integration lane; implementation steps; positive acceptance criteria; unhappy paths; tests first; real verification; >=95% changed-line and touched-module coverage; and `requiresDetailedPlan` only when needed.

Implementation-ready items dispatch directly to build. Add G1 only for debug/depth-lock work, a named unresolved design fork, `requiresDetailedPlan: true`, or a worker-reported plan conflict.

Decompose the mission into every genuinely disjoint lane. Never collapse a multi-surface mission into one "bounded" lane to shrink the roadmap; disjoint surfaces get disjoint lanes with disjoint ownership.

## 5. New-run governance

New-run governance is exactly:

1. `PROMPTS.txt` - exact append-only prompt blocks;
2. `ROADMAP.md` - canonical executable roadmap;
3. `GATELOG.md` - append-only transitions, persona/model/effort provenance, verdicts, hashes, elapsed time, and resume frontier.

Do not create new-run governance-only `BRIEF.md`, `PLAN.md`, `AGENTS.md`, `COVERAGE.md`, `BACKLOG.md`, `ANCHOR.md`, `bucketlist.md`, `intake.md`, `scope-map.md`, or per-angle scope files. Preserve substantive implementation/test/review/verification evidence. Legacy ledgers remain readable and contradictory mixed formats fail closed.

Governance lives at the run's governance root outside the mission target repository: `PROMPTS.txt`, `ROADMAP.md`, and `GATELOG.md` are never written into the target working tree and must never appear in its diff.

## 6. Compact pointer briefs

The first roadmap author stores the exact mission in `PROMPTS.txt`. Later briefs carry:

```text
MISSION POINTER: read the exact prompt ledger before acting; stop if its hash or byte length differs.
path=<PROMPTS.txt> hash=sha256:<64 hex> bytes=<UTF-8 byte length> nonce=<RUN-NONCE>
```

Workers verify path, hash, byte length, and nonce before acting. Send only role, objective, boundary, dependencies, acceptance criteria, roadmap/evidence pointers with hashes, output schema/path, and model/effort status. Do not paste the mission, transcript, full roadmap, doctrine, or prior adversarial reasoning. Preserve blind review.

## 7. Hierarchy and dispatch

Every worker is an installed, registered `ap-*` persona. Its custom-agent definition plus the dispatched task brief are its complete operating context. A worker must never load, invoke, or re-invoke the Autoprompt skill or start a nested Autoprompt run; it executes only its persona instructions a