UNPKG

aicf-core

Version:

Universal AI Context Format (AICF) - Enterprise-grade AI memory infrastructure with 95.5% compression and zero semantic loss

github.com/Vaeshkar/aicf-core

Vaeshkar/aicf-core

136 lines (108 loc) • 6.21 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
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
# Project File Access Patterns & Context > **🔒 PROTECTED FILE - Generated on 2025-10-05T17:30:44Z from Augment Data** > **Data Recovery:** Rebuilding lost context from original create-ai-chat-context project > **Source:** recentlyOpenedFiles.json (60+ tracked files) --- ## 📂 Active Development Areas (From File Access History) ### Core Project Structure ``` create-ai-chat-context/ ├── .ai/ # Human-readable knowledge base │ ├── conversation-log.md # Most accessed (primary history) │ ├── technical-decisions.md # Architecture decisions │ ├── next-steps.md # Current priorities │ ├── known-issues.md # Bug tracking │ └── architecture.md # System design ├── .aicf/ # AI-optimized formats (heavily accessed) │ ├── conversations.aicf # Compressed conversation history │ ├── index.aicf # Fast lookup metadata │ ├── tasks.aicf # Compressed task tracking │ ├── issues.aicf # Compressed issue tracking │ └── .meta # Project metadata ├── src/ # Core implementation │ ├── checkpoint-agent.js # (ARCHIVED - automated compression) │ ├── config.js # Configuration management │ ├── context-extractor.js # Context extraction utilities │ └── search.js # Search functionality ├── bin/cli.js # CLI entry point (frequently modified) └── docs/ # Documentation experiments └── aicf-evaluation-findings.md # Format evaluation research ``` --- ## 🔥 Most Frequently Accessed Files (From Augment Tracking) ### Primary Development Files: 1. **`.ai/conversation-log.md`** - Central conversation history (most critical) 2. **`bin/cli.js`** - CLI interface (frequent updates) 3. **`.aicf/conversations.aicf`** - AI-optimized conversation storage 4. **`src/checkpoint-agent.js`** - (Now archived - automated compression experiment) ### Active Experimentation Areas: 1. **`.aicf/checkpoint-queue/`** - Token compression testing (multiple test files) - `TEST-GPT4O-raw.aicf` - `checkpoint-20k-raw.aicf` - Multiple test result files with different models 2. **`archive/abandoned-automated-compression/`** - Moved failed experiments 3. **`docs/LOGIC_AGENT_ORCHESTRATOR_DESIGN.md`** - Architecture planning ### Configuration & Setup: - **`.env.local`** - Environment configuration (API keys) - **`.env.example`** - Template for environment setup - **`extract-warp-conversation.js`** - Cross-platform extraction (current work!) --- ## 📊 Development Pattern Analysis ### File Access Frequency Patterns: - **Daily access**: conversation-log.md, cli.js, .aicf files - **Weekly access**: technical-decisions.md, next-steps.md, architecture.md - **Experimental access**: checkpoint-queue files, test files - **Archive access**: Moved abandoned automation experiments ### Project Evolution Phases (From File Timeline): 1. **Initial CLI Development** - bin/cli.js, src/ files 2. **AI Format Experimentation** - .aicf/ directory creation 3. **Automated Compression Attempt** - checkpoint-agent.js (later archived) 4. **Manual Workflow Adoption** - Simplified .aicf approach 5. **Cross-Platform Integration** - extract-warp-conversation.js (current) --- ## 🔄 Active Development Context ### Current Focus Areas (From Recent File Access): 1. **Cross-Platform Extraction**: `extract-warp-conversation.js` 2. **Token Optimization**: Multiple .aicf test files 3. **Manual Workflow**: Simplified approach after abandoning automation 4. **Documentation**: Evaluation findings and design documents ### Abandoned Experiments (Moved to Archive): - **Automated compression agents** - Too complex, user preferred manual approach - **Complex checkpoint systems** - Overcomplicated token management - **SDK integrations** - @anthropic-ai/sdk, openai installs later removed ### Key Insights from File Patterns: - **Heavy .aicf experimentation** - Multiple test files indicate serious token optimization work - **Iterative CLI development** - Frequent bin/cli.js updates show active feature development - **Documentation-driven approach** - Many .md files indicate thorough documentation practices - **Cleanup and archiving** - Clear separation of working vs abandoned code --- ## 🎯 File Relationship Map ### Core Knowledge Flow: ``` conversation-log.md → conversations.aicf → index.aicf ↓ ↓ ↓ technical-decisions.md → technical-context.aicf → .meta ↓ ↓ ↓ next-steps.md → tasks.aicf → work-state.aicf ``` ### Development Workflow: ``` Local Development: Testing: Archive: ├── src/*.js → ├── test-*.js → └── archive/abandoned-* ├── bin/cli.js → ├── node bin/cli.js └── .ai/*.md → └── npx aic [cmd] ``` --- ## 🧠 Project Memory Context ### What File Patterns Tell Us: 1. **Token optimization was a major focus** - Extensive .aicf testing 2. **User prefers manual over automated** - Archived automation experiments 3. **Cross-platform integration current priority** - extract-warp-conversation.js 4. **Documentation is comprehensive** - Many .md files with detailed analysis ### Critical Files for Context Recovery: - `conversation-log.md` - Contains full project history - `.aicf/conversations.aicf` - Token-optimized conversation memory - `command-execution-history` - Development progression (we extracted this!) - `technical-decisions.md` - Architecture rationale --- **File Access Summary:** 60+ files tracked, indicating active, experimental development focused on AI memory optimization and cross-platform integration. The current work on `extract-warp-conversation.js` aligns perfectly with the multi-platform universal AI knowledge strategy we're implementing.