SlideRAG/env.project.example at main · iLearn-Lab/SlideRAG · 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
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
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
# ================================================================
# SlideRAG environment template (code-verified)
# This file only includes environment variables that are actually
# read by the project source code in client/ and rag_agent/.
# Copy to .env and fill values before running.
# ================================================================

# -------------------------------
# Core model provider settings
# Required for app.py / QQ runtime / Weixin runtime
# -------------------------------
OPENAI_API_KEY=your_openai_api_key
OPENAI_BASE_URL=https://api.your-provider.com/v1

# Text and vision models used by RAGAnything
TEXT_LLM_MODEL=gpt-4o-mini
VLM_MODEL=gpt-4o

# Agent provider for rag_agent loop: openai | anthropic
AGENT_PROVIDER=openai
AGENT_MODEL=gpt-4o

# Anthropic provider settings (only required when AGENT_PROVIDER=anthropic)
# If empty, runtime may fallback to OPENAI_API_KEY / OPENAI_BASE_URL.
ANTHROPIC_API_KEY=
ANTHROPIC_BASE_URL=

# Embedding model settings
EMBEDDING_MODEL=text-embedding-3-large
EMBEDDING_DIM=3072
EMBEDDING_MAX_TOKEN_SIZE=8192


# -------------------------------
# Document parser and retrieval
# -------------------------------
# Generic RAGAnything working directory (used by raganything/config.py)
WORKING_DIR=./rag_storage

PARSER=mineru
PARSE_METHOD=auto
ENABLE_IMAGE_PROCESSING=true
ENABLE_TABLE_PROCESSING=true
ENABLE_EQUATION_PROCESSING=true
RETRIEVE_TOP_K=20
RETRIEVE_CHUNK_TOP_K=20

# Advanced parser/context options (optional)
SUMMARY_LANGUAGE=English
ENABLE_HIDDEN_EXPANSION=true
HIDDEN_EXPAND_SCORE_THRESHOLD=0.55
HIDDEN_EXPAND_MAX_ITEMS_PER_PAGE=1
HIDDEN_EXPAND_USE_LLM=true
HIDDEN_EXPAND_MAX_CHARS=1200
SUPPORTED_FILE_EXTENSIONS=.pdf,.jpg,.jpeg,.png,.bmp,.tiff,.tif,.gif,.webp,.doc,.docx,.ppt,.pptx,.xls,.xlsx,.txt,.md
CONTEXT_WINDOW=1
CONTEXT_MODE=page
MAX_CONTEXT_TOKENS=2000
CONTENT_FORMAT=minerU

# -------------------------------
# Web app (client/app.py) paths
# -------------------------------
OUTPUT_DIR=./output
DOC_STORE_DIR=./uploaded_docs
WORKING_DIR_ROOT=./rag_storage_by_file
REGISTRY_PATH=./uploaded_docs_registry.json


# -------------------------------
# QQ runtime (client/qq_runtime.py)
# -------------------------------
QQ_ENABLED=false
QQ_APP_ID=
QQ_SECRET=
QQ_ALLOW_FROM=*
QQ_MSG_FORMAT=plain

# Optional startup target file for auto-ingest
QQ_TARGET_FILE=

# Runtime directories
QQ_UPLOADED_DOCS_DIR=./uploaded_docs
QQ_INGEST_OUTPUT_DIR=./output
QQ_RAG_WORKING_DIR=./rag_storage_by_qq_file
QQ_RUNTIME_STATE_DIR=./rag_storage_qq_runtime

# Startup ready notification
QQ_STARTUP_NOTIFY_ENABLED=true
QQ_STARTUP_NOTIFY_MESSAGE=rag agent is ready.
QQ_STARTUP_NOTIFY_CHAT_ID=


# -------------------------------
# Weixin runtime (client/weixin_runtime.py)
# -------------------------------
WEIXIN_ENABLED=true
WEIXIN_ALLOW_FROM=*
WEIXIN_BASE_URL=https://ilinkai.weixin.qq.com
WEIXIN_STATE_DIR=./rag_storage_weixin_runtime
WEIXIN_POLL_TIMEOUT=35

# Optional startup target file for auto-ingest
WEIXIN_TARGET_FILE=

# Runtime directories
WEIXIN_RAG_WORKING_DIR=./rag_storage_by_weixin_file
WEIXIN_RUNTIME_STATE_DIR=./rag_storage_weixin_runtime
WEIXIN_UPLOADED_DOCS_DIR=./uploaded_docs
WEIXIN_INGEST_OUTPUT_DIR=./output

# Startup ready notification
WEIXIN_STARTUP_NOTIFY_ENABLED=true
WEIXIN_STARTUP_NOTIFY_MESSAGE=agent is ready.
WEIXIN_STARTUP_NOTIFY_CHAT_ID=


# -------------------------------
# Feishu channel (rag_agent/channels/feishu.py)
# Used by feishu_smoke_test and custom runtime integration
# -------------------------------
FEISHU_ENABLED=true
FEISHU_APP_ID=
FEISHU_APP_SECRET=
FEISHU_ENCRYPT_KEY=
FEISHU_VERIFICATION_TOKEN=
FEISHU_ALLOW_FROM=*
# feishu (China) | lark (global)
FEISHU_DOMAIN=feishu

# Optional startup target file for auto-ingest
FEISHU_TARGET_FILE=

# Runtime directories
FEISHU_UPLOADED_DOCS_DIR=./uploaded_docs
FEISHU_INGEST_OUTPUT_DIR=./output
FEISHU_RAG_WORKING_DIR=./rag_storage_by_feishu_file
FEISHU_RUNTIME_STATE_DIR=./rag_storage_feishu_runtime

# Startup ready notification
FEISHU_STARTUP_NOTIFY_ENABLED=true
FEISHU_STARTUP_NOTIFY_MESSAGE=agent is ready.
FEISHU_STARTUP_NOTIFY_CHAT_ID=

# -------------------------------
# WhatsApp runtime (client/whatsapp/runtime.py)
# Requires a local WhatsApp bridge service.
# -------------------------------
WHATSAPP_ENABLED=false
WHATSAPP_ALLOW_FROM=*
WHATSAPP_BRIDGE_URL=ws://127.0.0.1:3001
WHATSAPP_BRIDGE_TOKEN=
WHATSAPP_RECONNECT_DELAY_S=5
WHATSAPP_SEND_RETRY_ATTEMPTS=3
WHATSAPP_SEND_RETRY_DELAY_MS=400
WHATSAPP_ACCEPT_GROUP_MESSAGES=true
WHATSAPP_REQUIRE_MENTION_IN_GROUP=true

# Optional startup target file for auto-ingest
WHATSAPP_TARGET_FILE=

# Runtime directories
WHATSAPP_UPLOADED_DOCS_DIR=./uploaded_docs
WHATSAPP_INGEST_OUTPUT_DIR=./output
WHATSAPP_RAG_WORKING_DIR=./rag_storage_by_whatsapp_file
WHATSAPP_RUNTIME_STATE_DIR=./rag_storage_whatsapp_runtime

# Startup ready notification
WHATSAPP_STARTUP_NOTIFY_ENABLED=true
WHATSAPP_STARTUP_NOTIFY_MESSAGE=agent is ready.
WHATSAPP_STARTUP_NOTIFY_CHAT_ID=

# -------------------------------
# Network and cache compatibility
# -------------------------------
# Local cache directory for tiktoken encoding files.
# Recommended in restricted network/offline environments to avoid startup failures.
TIKTOKEN_CACHE_DIR=./temp/tiktoken