yam/yam.manifest.json at main · 0kim0bos/yam · GitHub

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
{
  "name": "yam",
  "version": "2.4.0",
  "principles": [
    "Direction before execution.",
    "Start fast.",
    "Deepen deliberately.",
    "Basic direction fit and honest verification always.",
    "Honest by design.",
    "Replace or uninstall skills only when receipt ownership plus the complete regular-file inventory and content hashes prove they are yam-managed; preserve unproven user files by default.",
    "Heavy proof when risk or user intent calls for it.",
    "Token-aware by default.",
    "Context-reuse first.",
    "End with remaining tasks and fix-first items.",
    "When code, config, release metadata, documentation, or project artifacts changed, include a Study Note even if no yam skill was explicitly invoked.",
    "When the opt-in Study Note hook is enabled, check the final response at Codex Stop and request one bounded correction pass instead of relying only on prompt-time advice.",
    "Explain what changed, what role it has, where it runs or is read, what was wrong or missing when relevant, before/after behavior, expected behavior, and one useful syntax or structure insight in short non-specialist language.",
    "When relevant, include architecture hygiene: avoid dumping unrelated logic into page.tsx, one-off component CSS into global.css, or structured product data into broad DB jsonb blobs when better project boundaries exist.",
    "Use handoff notes for fix-first items, remaining tasks, recommended next direction, requirement coverage, blocker kind, failure cause, recovery hint, safe retry, evidence level, and owner scope when useful.",
    "Use bounded heavy verification for serious claims, with explicit stop conditions instead of endless proof loops."
  ],
  "defaultHooks": false,
  "routes": [
    {
      "id": "quick",
      "stage": "v0.3",
      "purpose": "Fast scoped implementation, small fixes, and quick error scans with focused verification."
    },
    {
      "id": "ueye",
      "stage": "v0.4",
      "purpose": "Tight UI/UX/design implementation and visual review with visual provenance, local asset integrity, revision history, reference-read proof, visual comparison, design quality review, and P0-P3 issue tracking."
    },
    {
      "id": "question",
      "stage": "v0.1",
      "purpose": "Very-light direct Q&A for stable or local questions without implementation."
    },
    {
      "id": "scout",
      "stage": "v0.2",
      "purpose": "Investigation, wide search, evidence-led source mapping, blocked-source fallback, opposition pass, and recommendation without code changes by default."
    },
    {
      "id": "deep",
      "stage": "v0",
      "purpose": "Single-agent heavy verification by explicit request, including runtime/tmux/browser/process proof when needed."
    },
    {
      "id": "mission",
      "stage": "v0.3",
      "purpose": "Explicit real-subagent/team execution route for approved plans, read-only reviewer/doctor contracts, per-thread receipts, aggregate completion gate, patch envelopes, rollback hints, cross-verification, doctor scan, and final proof."
    }
  ]
}