GodsDevProject commited on
Commit
bf01fe5
Β·
verified Β·
1 Parent(s): 4394fda

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +58 -175
README.md CHANGED
@@ -13,215 +13,98 @@ short_description: 'FOIA DECLASSIFIED DOCUMENTS SEARCH '
13
 
14
  # πŸ›οΈ Federal FOIA Intelligence Search
15
 
16
- **Public FOIA Electronic Reading Rooms β€’ Link-Out Only β€’ Court-Aware**
17
 
18
- A Hugging Face–hosted research, journalism, and legal-support tool for discovering, organizing, and exporting **public U.S. government FOIA materials** from official electronic reading rooms.
19
 
20
- This application **does not scrape, host, or redistribute documents**.
21
- All results are **direct links to official government FOIA libraries**.
22
 
23
  ---
24
 
25
- ## πŸ” Core Capabilities
26
 
27
- ### βœ” Federated FOIA Search (LIVE)
28
- Search across multiple U.S. government FOIA libraries simultaneously.
29
-
30
- **LIVE Agencies (Public & Safe):**
31
- - CIA β€” FOIA Electronic Reading Room
32
- - FBI β€” The Vault
33
- - DOJ β€” FOIA Library
34
- - DHS β€” FOIA Reading Room
35
- - State Department β€” FOIA Search
36
- - GSA β€” FOIA Library
37
- - NSA β€” FOIA Reading Room
38
-
39
- All results link directly to official government sources.
40
 
41
  ---
42
 
43
- ### πŸ§ͺ Extended Coverage (Clearly Labeled STUBS)
44
- Optional **non-exportable indicators** for agencies where automated access may be restricted or ambiguous.
45
-
46
- - DIA
47
- - NGA
48
- - NRO
49
- - TEN-CAP
50
- - AATIP
51
- - SAP / Special Activities
52
 
53
- **STUB results:**
54
- - ❌ No URLs
55
- - ❌ No exports
56
- - ❌ No PDFs
57
- - ❌ No citations
58
-
59
- This separation is a **deliberate compliance safeguard**.
60
 
61
  ---
62
 
63
- ### πŸ“„ PDF Thumbnail Gallery
64
- For results that link directly to `.pdf` files:
65
-
66
- - Inline iframe preview (HF-safe)
67
- - Action buttons:
68
- - **View**
69
- - **Download**
70
- - **Share** (device-native share API if supported)
71
- - **Ask AI** (safe placeholder β€” no ingestion)
72
 
73
- > PDFs are never downloaded, cached, or stored by the app.
 
 
 
 
 
 
 
 
74
 
75
  ---
76
 
77
- ### πŸ—‚ Journalist ZIP Export (LIVE Only)
78
- Generates a ZIP package for editorial or investigative workflows.
79
 
80
- **Contents:**
81
- - `README.txt` β€” scope & disclaimer
82
- - `citations.txt` β€” Bluebook-ready citations
83
- - `links.csv` β€” agency, title, URL, timestamp
84
- - `pdf_links.txt` β€” direct PDF URLs (no files)
85
 
86
- βœ” Public sources only
87
- βœ” No redistribution
88
- βœ” No STUB data
 
 
 
 
 
89
 
90
- ---
91
-
92
- ### 🌐 Public Shareable Result Pages
93
- Generate static, shareable result summaries containing:
94
- - Agency + title
95
- - Direct source links
96
- - Bluebook citations
97
- - Citation hash for integrity verification
98
 
99
  ---
100
 
101
- ### βš–οΈ Court-Aware Features
102
- Each LIVE result includes:
103
- - SHA-256 citation hash
104
- - Bluebook citation
105
- - Retrieval timestamp
106
 
107
- Supports:
108
- - Litigation appendices
109
- - Exhibit preparation
110
- - Evidentiary traceability
 
111
 
112
  ---
113
 
114
- ### 🧠 Semantic Mode (Opt-In)
115
- - FAISS + SentenceTransformers (optional)
116
- - Metadata embeddings only
117
- - Disabled by default
118
- - Auto-disabled if dependencies unavailable
119
 
120
- ---
 
 
 
 
121
 
122
- ### πŸ“Š Visual Analytics
123
- - Entity / domain frequency graphs
124
- - Retrieval timeline charts
125
 
126
  ---
127
 
128
- ### πŸ“ FOIA Request Generator
129
- Generate a fillable FOIA request PDF using:
130
- - Requester name
131
- - Description of records
132
- - Agencies surfaced in LIVE results
133
 
134
- ---
135
-
136
- ## 🧱 What This App Does NOT Do
137
-
138
- ❌ No scraping
139
- ❌ No crawling behind authentication
140
- ❌ No hosting or redistributing documents
141
- ❌ No classified access
142
- ❌ No private datasets
143
- ❌ No surveillance or tracking
144
 
145
  ---
146
 
147
- # πŸ›‘οΈ TRUST & SAFETY ADDENDUM (HF REVIEWERS)
148
-
149
- ### Compliance Summary
150
- This Space is intentionally designed to comply with:
151
- - Hugging Face Spaces policies
152
- - U.S. FOIA public-access norms
153
- - Journalism ethics standards
154
-
155
- ### Safety Controls
156
- - **Link-out only** (no content ingestion)
157
- - **Explicit STUB labeling**
158
- - **Export gating (LIVE-only)**
159
- - **No PDF storage**
160
- - **No background crawling**
161
- - **User-initiated queries only**
162
-
163
- ### Risk Mitigation
164
- - No robots.txt violations (manual user navigation)
165
- - No automated retrieval of sensitive systems
166
- - No claims of completeness or authority
167
-
168
- This Space functions as a **research navigation and citation tool**, not a data collection system.
169
-
170
- ---
171
-
172
- # βš–οΈ LEGAL REVIEW MEMO (NON-BINDING)
173
-
174
- ### Scope
175
- This application operates entirely within:
176
- - Public FOIA Electronic Reading Rooms
177
- - User-directed navigation
178
-
179
- ### Key Legal Characteristics
180
- - No republication of copyrighted works
181
- - No derivative document creation
182
- - No alteration of government records
183
- - No implied agency endorsement
184
-
185
- ### Litigation Safety
186
- - Citation hashes provide integrity verification
187
- - Source URLs remain authoritative
188
- - Export artifacts include disclaimers
189
-
190
- This tool supports lawful research and reporting, not legal conclusions.
191
-
192
- ---
193
-
194
- # πŸ“° JOURNALIST ONBOARDING GUIDE
195
-
196
- ### Typical Workflow
197
- 1. Enter investigative topic
198
- 2. Review LIVE agency coverage
199
- 3. Examine PDF previews
200
- 4. Export journalist ZIP
201
- 5. Generate FOIA follow-up request
202
- 6. Share result summary with editors
203
-
204
- ### Best Practices
205
- - Always open documents at the source
206
- - Use citation hashes for verification
207
- - Treat STUB indicators as leads only
208
- - File FOIA requests directly with agencies
209
-
210
- ### Ethical Use
211
- - Attribute sources accurately
212
- - Avoid implying classified access
213
- - Verify context before publication
214
-
215
- ---
216
-
217
- ## ⚠️ Disclaimer
218
- This tool:
219
- - Is not affiliated with any U.S. government agency
220
- - Does not guarantee completeness
221
- - Does not provide legal advice
222
-
223
- ---
224
 
225
- **Built for transparency.
226
- Designed for accountability.
227
- Safe by construction.**
 
13
 
14
  # πŸ›οΈ Federal FOIA Intelligence Search
15
 
16
+ **Public FOIA Electronic Reading Rooms β€” Link-Out Search Only**
17
 
18
+ This Hugging Face Space provides a **federated discovery interface** for official U.S. Government Freedom of Information Act (FOIA) Electronic Reading Rooms.
19
 
20
+ ⚠️ **This application does not scrape, crawl, mirror, store, or redistribute documents.**
21
+ All results link directly to authoritative government sources.
22
 
23
  ---
24
 
25
+ ## 🎯 Purpose
26
 
27
+ - Accelerate public-interest research
28
+ - Support journalism, litigation prep, and academic inquiry
29
+ - Provide court-ready citations and documentation workflows
30
+ - Enable **explicitly gated AI analysis** for public documents
 
 
 
 
 
 
 
 
 
31
 
32
  ---
33
 
34
+ ## βœ”οΈ Core Guarantees
 
 
 
 
 
 
 
 
35
 
36
+ - πŸ”— **Link-out only** to official FOIA libraries
37
+ - πŸ“„ **No document hosting or redistribution**
38
+ - 🧾 **Bluebook-formatted citations**
39
+ - 🧠 **AI analysis is explicit opt-in**
40
+ - βš–οΈ **LIVE results only are exportable**
41
+ - 🚫 **STUB coverage is informational only**
 
42
 
43
  ---
44
 
45
+ ## πŸ›οΈ Live FOIA Sources
 
 
 
 
 
 
 
 
46
 
47
+ | Agency | Official Source |
48
+ |------|----------------|
49
+ | CIA | cia.gov/readingroom |
50
+ | FBI | vault.fbi.gov |
51
+ | DOJ | justice.gov/foia |
52
+ | DHS | dhs.gov/foia |
53
+ | State Dept | foia.state.gov |
54
+ | GSA | gsa.gov |
55
+ | NSA | nsa.gov |
56
 
57
  ---
58
 
59
+ ## 🧠 AI & Semantic Features (Opt-In)
 
60
 
61
+ AI functionality is **disabled by default**.
 
 
 
 
62
 
63
+ When enabled by the user:
64
+ - PDFs are processed **only when explicitly selected**
65
+ - AI analysis is **user-initiated**
66
+ - Outputs include:
67
+ - Disclosure footer
68
+ - Integrity hash
69
+ - Source citation
70
+ - AI output **is not a primary source** and **not legal advice**
71
 
72
+ Optional FAISS semantic indexing operates on **metadata only** unless PDF analysis is explicitly requested.
 
 
 
 
 
 
 
73
 
74
  ---
75
 
76
+ ## πŸ—‚οΈ Exports & Research Tools
 
 
 
 
77
 
78
+ - 🧾 Bluebook citations
79
+ - πŸ“¦ Journalist ZIP (links + citations only)
80
+ - πŸ“„ FOIA request generator (PDF)
81
+ - πŸ›οΈ Litigation appendix generator
82
+ - πŸ“Š Entity graphs & timelines
83
 
84
  ---
85
 
86
+ ## πŸ” Trust & Safety
 
 
 
 
87
 
88
+ - No scraping or crawling
89
+ - No robots.txt bypass
90
+ - No credentialed systems
91
+ - No personal data collection
92
+ - Stateless execution
93
 
94
+ See:
95
+ - `TRUST_SAFETY.md`
96
+ - `LEGAL_MEMO.md`
97
 
98
  ---
99
 
100
+ ## πŸš€ Deployment
 
 
 
 
101
 
102
+ Designed for **Hugging Face Spaces (Gradio)**
103
+ No API keys required.
 
 
 
 
 
 
 
 
104
 
105
  ---
106
 
107
+ ## πŸ“œ Disclaimer
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
108
 
109
+ This tool assists discovery and analysis of public records.
110
+ Always verify information against the original government source.