Help Center

User Guide

Everything you need to get the most out of the Intelligence Archive.

Browsing the Archive

The Browse page displays all published documents as a grid of thumbnail cards spanning 1939–2007. Each card shows the document ID, date, title, type, and classification level.

At the top of the Browse page, a row of Featured Documents highlights items of particular research interest — documents selected for their historical significance across the archive's major collections. These are a useful starting point for new visitors.

The sidebar on the left (desktop) or the Scopes toggle (mobile) organises the archive into three focus areas, each collapsible:

  • CIA Albania Operations — covert operations in Albania from the late 1940s through the 1990s, including the Enver Hoxha scope.
  • OSS to Early CIA — OSS wartime records and the institutional transition through SSU and CIG to the founding of the CIA in 1947.
  • U.S. Intelligence & Former Nazi Personnel — CIA recruitment of former Nazi and SS personnel, U.S.–BND relations, and German intelligence activities in WWII.

Click a focus area header to select or deselect all scopes within that group at once. Expand the group with the chevron to toggle individual scopes. Document counts are shown on each scope tile.

Click Filters to open the filter panel. You can narrow results by document type, operation scope, classification level, and date range. The date pickers open a calendar bounded to the archive period — no value means no filter is applied.

A green Downloaded badge appears on any document you have already downloaded, so you never pay twice for the same file. Results are paginated in sets of 24; use the Previous / Next controls to navigate.

Tip: The search box on the Browse page searches metadata only (title, ID, description, entities, keywords). For full-text OCR search inside documents, use Advanced Search.

AI Research Assistant

Per-Document Analysis

Every document detail page has an AI Assistant tab. Ask any question about the document and the assistant will answer using the document's transcribed text, always providing the passage it drew from so you can verify the source.

Query quotas are enforced per account:

TierMonthly limitDaily limit
Free5 queries3 queries
Researcher300 queriesNo daily cap
ScholarUnlimitedNo daily cap

Your remaining quota is shown as AI queries remaining this month on the document page. Quotas reset on the first of each month.

Cross-Document Folder Analysis

Every research folder has an AI Analysis tab. Open a folder from My Workspace, click the AI Analysis tab, and ask the AI to analyse all documents in that folder simultaneously. The AI identifies patterns and connections across people, organisations, locations, cryptonyms, and recurring themes — surfacing relationships that would take hours to find manually.

Example: create a folder called Albania Coup Planning, add 4–6 documents covering the same operation, then ask 'What cryptonyms appear most frequently across these documents, and what do they refer to?' or 'Which individuals appear in multiple documents and in what roles?'

Folder AI Analysis queries draw from the same monthly quota as per-document queries. Folders that share a common operation, time period, or subject produce the most coherent analysis — random collections yield less useful results.

Tip: OCR quality affects both analysis modes. When a document's condition is flagged as Poor OCR quality, AI results for that document may be incomplete or inaccurate. Check the Document Condition field on the Document Detail page before relying on AI output for degraded records.

Document Intelligence Layer

Every document in the archive has been automatically processed to extract structured intelligence beyond basic metadata. These fields appear on the Document Detail page and directly power the entity filters in Advanced Search.

Cryptonyms — CIA code names found in the document (e.g., BGFIEND, FIEND, QKIVORY, TPFEELING). No other public archive automatically identifies and surfaces cryptonyms. They are directly searchable in Advanced Search, connecting documents across operations.
Acronyms with Definitions — Intelligence-specific acronyms are parsed and defined (e.g., COP — Chief of Operations; ADPC — Assistant Deputy for Policy Coordination). This removes a significant language barrier for non-specialist researchers.
Collective Terms — Informal collective references found in the document (e.g., 'British', 'Albanians') identified to aid entity-based searches.
Dominant Themes — AI-extracted thematic tags (e.g., 'inter-agency coordination', 'operational security') that support topic-based discovery across the archive.
Key Entities — People (with roles) and Organisations (with type) extracted from each document, linked to the entity filter in Advanced Search.
Keywords — Specific terms extracted from the document text relevant to research queries.
Tip: Cryptonyms, themes, acronyms, and entity names are all indexed as dedicated search fields — not buried in OCR text. Searching BGFIEND returns every document where that cryptonym was identified, even if the transcribed text is too degraded to contain it. This makes intelligence-layer fields searchable with the same precision as structured filters.

Downloading Documents & Page Credits

Free tier — you can view a watermarked PDF preview of every document. The preview is limited to the first 5 pages. The watermark contains your name and the download date for attribution purposes.

Researcher and Scholar tiers — you can download the full, unmarked PDF. Downloads are debited in page credits: a 12-page document costs 12 credits.

Re-downloads are free. Once you have downloaded a document, you can download it again at any time without spending credits. The green Downloaded badge on browse and search cards marks documents you own.

Page credit packs — if you exhaust your monthly subscription credits you can purchase additional packs of 100 pages for $10. Packs never expire and are consumed after your monthly subscription credits are exhausted.

AI Query Packs — if you exhaust your monthly AI query quota, you can purchase additional packs of 100 queries for $10. Query packs never expire and are consumed after your monthly subscription queries are exhausted.

Tip: Check the page count shown on the document detail page before downloading if you are running low on credits.

Research Folders

Folders let you organise documents into named collections for ongoing research projects — similar to a physical document binder.

Create a folder from the Folders page in My Workspace. Add documents using the folder icon on any browse card, search result card, or document detail page. You can add up to 10 documents at a time using the multi-select checkboxes on Browse or Search.

The folder counter — shown as 1/10 for Researcher-tier users — updates in real time as you create or delete folders, confirming your usage against your plan limit.

Each folder includes an AI Analysis tab that enables cross-document analysis across all documents in that folder simultaneously. See the AI Research Assistant section for full documentation of this feature.

TierFoldersDocs per folder
Free210
Researcher10Unlimited
ScholarUnlimitedUnlimited

Saved Searches

After running a search, click the Save button in the search form to give the search a name and store it in your workspace. Saved searches remember all active filters — full-text query, entities, date range, classification, type, and scope.

Access your saved searches from My Workspace → Saved Searches. Click any saved search to re-run it instantly with the current archive data.

Tip: Use saved searches to monitor recurring research topics — re-run them periodically as new documents are added to the archive.

Cite & Export

The Cite & Export button on every Document Detail page generates a formatted citation or exports a document reference directly from the archive.

Available citation formats:

Chicago, MLA, APA — available on Researcher and Scholar tiers.
Basic citation — available on the Free tier.

To cite a document: open the Document Detail page → click Cite & Export → select your format → copy the citation or download it as a text file.

Export Chat to Word — Researcher and Scholar users can export an AI chat session from the Document Detail AI Assistant tab to a Word-compatible document. Open the AI Assistant tab on any document, complete your research session, then click Export Chat to Word.

Tip: Citations include the document's NARA locator, classification level, and document date — all fields required for academic citation of declassified government records.

Subscription & Billing

Intelligence Archive offers three tiers. You can upgrade, downgrade, or cancel at any time from Account & Billing in My Workspace.

FeatureFreeResearcherScholar
PDF accessWatermarked preview (first 5 pages)Full downloadFull download
Page credits/month0300 pagesUnlimited
AI queries/month5300Unlimited
Daily AI cap3/dayNoneNone
Folders210Unlimited
Docs per folder10UnlimitedUnlimited
Saved searches21010

Add-on purchases — available to all paid tiers from Account & Billing:

Page credit packs — 100 pages · $10 · never expire · consumed after monthly subscription credits
AI Query Packs — 100 queries · $10 · never expire · consumed after monthly subscription queries

Cancellation takes effect at the end of your current billing period — you keep access to your tier until then. Page credit packs and AI Query Packs are non-refundable and never expire.

For group or volume licensing, contact us at hello@intelarchive.com.

Frequently Asked Questions

What happens when I run out of page credits?

You will see a message indicating how many pages are required vs. how many you have remaining. You can purchase a page credit pack (100 pages, $10) from the Account & Billing page, or upgrade to a higher subscription tier. Downloads are blocked until credits are available.

Can I re-download a document I've already paid for?

Yes. Once you have downloaded a document, every subsequent download is free — no credits are deducted. The green Downloaded badge in browse and search results identifies documents you already own.

Do page credit packs expire?

No. Page credit packs never expire. They remain on your account until consumed and carry over if you change or cancel your subscription. Pack credits are consumed after your monthly subscription credits each billing cycle.

What does the AI assistant have access to?

The AI assistant has two modes. On a Document Detail page it has access to the OCR-transcribed text of that specific document only. In a research folder, the AI Analysis tab analyses all documents in the folder simultaneously, identifying cross-document patterns in people, organisations, locations, cryptonyms, and themes. In both modes, answers include source passages so you can verify them directly.

Can I share a search with a colleague?

Yes. The URL on the Search page encodes all active filters. Copy the URL from your browser and share it — anyone who opens it will see the same search with all the same filters pre-applied. Note that search results depend on their own subscription tier.

How do I cancel my subscription?

Go to My Workspace → Account & Billing and click Manage Subscription. You will be taken to the Stripe billing portal where you can cancel. Your access continues until the end of the current billing period.

What file format are downloaded documents?

All documents are PDF files, drawn directly from the original scanned records. File sizes vary depending on the number of pages and scan resolution. The page count is shown on every document card and detail page before you download.

Why is a document's text missing or garbled?

OCR quality depends on the condition of the original scanned document. Heavily redacted, water-damaged, or poor-quality scans may produce incomplete or inaccurate transcriptions. The original PDF is always the authoritative source. When AI results seem incomplete, check the Document Condition field on the Document Detail page.