File size: 4,970 Bytes
5e1a091
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
041a9c9
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
// The 12 harness-ready sections — shared by server (seeding, reset) and client (done detection).
export const SECTION_TEMPLATES = [
  {
    key: "meta",
    title: "Header & Metadata",
    hint: "Status, owner, reviewers, last-updated, and links to the PRD and Scope. Pin real dates and names — no TODOs.",
    starter: `| Field | Value |\n|---|---|\n| Status | DRAFT |\n| Owner | |\n| Reviewers | |\n| Last updated | |\n| Implements | PRD vX, Scope vX |\n| Repo | |`,
  },
  {
    key: "context",
    title: "Context & Background",
    hint: "One paragraph: why this exists, what exists today (link it), what breaks. A stranger must not need the conversation.",
    starter: "**Why:** \n\n**What exists today:** \n\n**What breaks / what's demanded:** (cite FR ids from the PRD)",
  },
  {
    key: "arch",
    title: "Architecture Overview",
    hint: "One diagram a stranger AND an agent can navigate — Mermaid preferred. Pin the exact stack versions below it.",
    starter: "```mermaid\nflowchart LR\n  Client[Client / UI] --> API[API Layer]\n  API --> Core[Domain Core]\n  Core --> DB[(Datastore)]\n```\n\n**Stack (pinned):** language , framework , DB ",
  },
  {
    key: "layout",
    title: "Directory / Module Layout",
    hint: "Agents navigate by path. Give the target tree up front so every task lands in a known place.",
    starter: "```\nrepo/\n├── app/\n│   ├── api/     # route handlers, no business logic\n│   ├── core/    # domain logic\n│   └── models/  # data models & schemas\n└── tests/       # mirrors app/\n```",
  },
  {
    key: "components",
    title: "Component Contracts",
    hint: "Per component: responsibility, typed interface, dependencies, and failure modes — what breaks, who retries, what the user sees.",
    starter: "### Component A\n- **Responsibility:** \n- **Interface:** signature(args: Type) -> Type\n- **Depends on:** \n- **Failure mode:** what happens when it breaks, who retries, what the user sees",
  },
  {
    key: "data",
    title: "Data Model & API Contracts",
    hint: "Schemas in full — types, nullability, indexes. Contracts get frozen and versioned; include migration + rollback strategy.",
    starter: "```sql\nCREATE TABLE example (\n  id TEXT PRIMARY KEY,\n  created_at TIMESTAMPTZ NOT NULL DEFAULT now()\n);\n```\n\n**API contract:** OpenAPI/JSON Schema at  — frozen.\n**Events:** names, payloads, delivery semantics.",
  },
  {
    key: "orchestration",
    title: "Orchestration (Build Harness)",
    hint: "How the build is orchestrated: master plan, atomic task schema (id/reads/writes/acceptance/blocked_by), the loop, state file, human checkpoints.",
    starter: "- **Master plan:** PLAN.md — numbered milestones, each ends demonstrable\n- **Task schema:** id, reads, writes, acceptance command, blocked_by\n- **Loop:** spec → decompose → generate → verify → commit or block\n- **State:** STATE.json with todo / in_progress / done / blocked\n- **Human checkpoints:** PR review per milestone; manual review for ",
  },
  {
    key: "nfr",
    title: "Non-Functional Requirements",
    hint: "Every -ility paired with how it's verified. \"p95 < 200 ms\", not \"fast\".",
    starter: "| Dimension | Requirement | How verified |\n|---|---|---|\n| Performance | p95 <  ms | |\n| Reliability | | |\n| Security | | |\n| Observability | | |",
  },
  {
    key: "testing",
    title: "Testing Strategy & Gate",
    hint: "Unit / integration / E2E mapped to story ids. Define the single CI command that is the definition of done.",
    starter: "- **Unit:** coverage target , exclusions \n- **Integration:** real vs mocked \n- **E2E:** maps to US ids \n- **Gate:** CI green on command before any task is done",
  },
  {
    key: "decisions",
    title: "Decision Log (ADRs)",
    hint: "Record why, not just what: context, options, decision, consequences. Prevents re-litigating choices.",
    starter: "| Decision | Alternatives | Why chosen | ADR |\n|---|---|---|---|\n|  |  |  | ADR-001 |",
  },
  {
    key: "risks",
    title: "Risks & Open Questions",
    hint: "Every open question gets an owner and a resolve-by date. Nothing open on the critical path.",
    starter: "- **Risk:**  → Mitigation: \n- **Open question:**  — Owner: , resolve by: ",
  },
  {
    key: "tasks",
    title: "Milestone → Task Map",
    hint: "The export surface: this table IS the harness's input. Task id, spec section, FR ref, deliverable files, acceptance command.",
    starter: "| Task | Spec section | FR | Deliverable | Acceptance check |\n|---|---|---|---|---|\n| T-01 | §  | FR- |  | command passes |",
  },
];

// A section counts as done when it holds more than the bare scaffold.
export const isSectionDone = (content) => {
  const c = (content ?? "").replace(/\s+/g, " ").trim();
  if (c.length < 60) return false;
  for (const t of SECTION_TEMPLATES) {
    if (c === t.starter.replace(/\s+/g, " ").trim()) return false;
  }
  return true;
};