-
Notifications
You must be signed in to change notification settings - Fork 109
Expand file tree
/
Copy pathversion.json
More file actions
96 lines (96 loc) · 6.69 KB
/
version.json
File metadata and controls
96 lines (96 loc) · 6.69 KB
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
86
87
88
89
90
91
92
93
94
95
96
{
"version": "3.5.0",
"lastUpdated": "2026-04-01T00:00:00Z",
"schemaVersion": "1.0",
"changelog": [
{
"version": "3.5.0",
"date": "2026-04-01",
"type": "minor",
"summary": "Meta-system update. Bootstrap optimized from 6 files (~17,800 tokens) to 5 files (~14,600 tokens): version.json moved to sleeping memory. AGENTS.md trimmed to persona + pointers. thinking.mdc stripped to 7 always-on coaching lenses (unique value only). Added .github/copilot-instructions.md as VS Code + Copilot bootstrap layer. ORCHESTRATION.md: company context routing guard, intent disambiguation, mid-conversation transition enforcement, Claude Code bootstrap ref. MEMORY.md wake tables updated. New frameworks: 10-Things Workshop (2.3.8), Documentation Standards (2.4.10). Setup files relocated from 01-Company-Context/ to docs/platform-setup.md. Eval system: agent behavior guide and scenarios updated."
},
{
"version": "3.4.0",
"date": "2026-02-27",
"type": "minor",
"summary": "Added a lightweight onboarding cluster under 01-Company-Context/1.3-Onboarding (Top 10 guide, metrics glossary, acronyms quick-reference) to make new-PM setup faster. Reworked the Product Judgment Test to be script-free by adding a transparent brier-calculation doc and a spreadsheet calculator, updating README/dashboard, and removing the Python helper script. Introduced a concrete Context Health table in 01-Company-Context/CONTEXT-HEALTH.md and wired it into MEMORY and ORCHESTRATION so the agent can offer gentle freshness checks when company/initiative docs feed into real decisions."
},
{
"version": "3.3.0",
"date": "2026-02-20",
"type": "minor",
"summary": "Refined README opening for stronger PM pain clarity and concise positioning. Added lightweight guidelines wiring by adding docs/guidelines.md wake triggers in MEMORY and reminder behavior in ORCHESTRATION. Standardized terminology to full forms (organization/communication), renamed 01-Company-Context/1.2-Organization-Survival to 1.2-Organization-Survival and updated links, and aligned agent naming to PM Brain Coach across AGENTS/setup/rules docs."
},
{
"version": "3.2.0",
"date": "2026-02-16",
"type": "minor",
"summary": "Added politics-coach skill, stakeholder avatars framework, and Organization-Survival company-context layer (power, alliances, stakeholder games, red flags). Updated ORCHESTRATION and MEMORY to route politics-related queries through avatars and Organization-Survival docs, and wired initiative stakeholders/ docs to this political layer. Normalized language for organization and communication across core docs."
},
{
"version": "3.1.0",
"date": "2026-02-16",
"type": "minor",
"summary": "Human docs moved to docs/ (setup, guidelines, architecture, credits, agent-manifest). Added Design Principles and Naming conventions to docs/architecture.md. README restructured as single entry point for humans, agents, and forkers. MEMORY.md wake trigger for repo structure/architecture topics. ORCHESTRATION.md: check filesystem before asking user about context. Renamed BRIER-CALCULATION.md to brier-calculation.md. All cross-references updated."
},
{
"version": "3.0.0",
"date": "2026-02-13",
"type": "major",
"summary": "Platform-agnostic core and context rot prevention. AGENTS.md and ORCHESTRATION.md no longer reference .cursor/ paths directly; all platform-specific paths now routed through MEMORY.md. Added Context Health section to ORCHESTRATION.md (heuristic checkpointing, re-anchoring, error recovery). Created CLAUDE.md for Claude Code. Added checkpoints/ folder for session state persistence. Updated SETUP.md with multi-platform setup guide."
},
{
"version": "2.2.0",
"date": "2026-02-13",
"type": "minor",
"summary": "Added communication style / voice layer. Agent now speaks with direct, experience-backed, prose-first personality. New voice.mdc (alwaysApply). Updated Core Persona in AGENTS.md and state-specific personas in 0-start-here-product-thinking.md. DevX===UX recurring theme added."
},
{
"version": "2.1.0",
"date": "2026-02-04",
"type": "minor",
"summary": "Added eval test generator and seeded test cases; introduced specialized skills (discovery-research, strategy-planning, stakeholder-management); added common slash commands; wired optional post-conversation eval logging hook; updated MEMORY and eval docs."
},
{
"version": "2.0.0",
"date": "2026-02-03",
"type": "major",
"summary": "Simplified orchestration: all agent decision logic now in ORCHESTRATION.md. Explicit state machine with clear transitions. AGENTS.md slimmed to persona and pointer to ORCHESTRATION. Sleeping memory manifest (MEMORY.md) added for context loading (company, initiatives, research, rules, skills)."
},
{
"version": "1.1.0",
"date": "2026-02-03",
"type": "minor",
"summary": "Product sense is now an integrated, first-class part of the repo: the agent orchestrates three modes (product_sense, execution_mode, meta_reflection), enforces braindump-before-structure, and uses a two-level evaluation system plus \"For Agents\" sections in 14 frameworks to guide user thinking and artifact quality end-to-end."
},
{
"version": "1.0.0",
"date": "2026-02-03",
"type": "initial",
"summary": "Initial version with complete framework library (14 frameworks), agent instructions (AGENTS.md), eval system (Level 1 & Level 2), 6 rules, 1 skill, and architecture documentation"
}
],
"breakingChanges": [
"Bootstrap set reduced from 6 files to 5: AGENTS.md, ORCHESTRATION.md, voice.mdc, thinking.mdc, thinking.personal.mdc. version.json is now sleeping memory (woken on demand via MEMORY.md), not bootstrap",
"thinking.mdc stripped to coaching lenses only — persona, braindump rules, core expectations, context sources, complex tradeoffs sections removed (canonical homes: AGENTS.md, ORCHESTRATION.md)",
".github/copilot-instructions.md is now the explicit VS Code + GitHub Copilot bootstrap layer for loading the always-on set before first response",
"Setup files moved from 01-Company-Context/SETUP.md and SETUP-QUICK-REFERENCE.md to docs/platform-setup.md"
],
"structure": {
"frameworks": 14,
"rules": 7,
"skills": 5,
"evalSystem": true,
"orchestrationFile": true,
"memoryManifest": true,
"platformAdapters": ["CLAUDE.md"],
"contextHealth": true,
"checkpoints": true,
"agentInstructions": {
"states": ["product_sense", "execution_mode", "meta_reflection", "conversation"],
"evalCheckpoints": 4,
"forAgentsSections": 14
},
"versionManagement": true
}
}