Google Cloud · Gemini Enterprise

Enterprise Knowledge Bases on Gemini Enterprise: Permission-Aware, Fresh and Measurable

A knowledge assistant earns trust in three ways: it never shows people what they should not see, it answers from the current version of the truth, and its quality is a number rather than an impression. This is how we build that on Gemini Enterprise — connectors, ACL sync, ingestion, custom RAG agents, grounding, freshness, lifecycle and evaluation.

TL;DR

Most enterprise knowledge assistants fail for the same reasons: they show people documents they should not see, quote policies that were replaced months ago, and nobody can say whether they are getting better or worse. Gemini Enterprise provides a strong base — indexed and federated connectors, permissions-aware answers and a customizable RAG agent. Success depends on the work around it: choosing indexed or federated per source, keeping ACLs in sync, parsing documents properly, attaching ownership and lifecycle to content, setting freshness SLOs, deciding when to add a custom ADK agent on RAG Engine or Vector Search, and measuring retrieval and answers with a golden set, online evaluation and Agent Simulation.

Gemini Enterprise on Google Cloud · a five-part series
1. Reference architecture · 2. Enterprise knowledge bases · 3. The agentic harness · 4. Real-time context · 5. Agentic L1 resolution

Why enterprise knowledge bases disappoint

The pilot usually goes well: a few hundred curated files and a friendly audience. Then scope widens to the real estate — thousands of SharePoint sites, a decade of Confluence, ServiceNow knowledge, shared drives of scanned PDFs — and three problems appear that have nothing to do with the model.

  • Permissions. An assistant that can read everything will eventually summarize a salary review, an M&A folder or a disciplinary file for someone who was never meant to see it. One incident like that ends the program.
  • Freshness and truth. The travel policy has three versions on three sites. The one the assistant finds first is the 2023 draft. The answer is well written, cited and wrong.
  • No measurement. When a business sponsor asks whether answers are good, the honest reply in most programs is “people seem to like it.” Without a golden set, every change is a guess and users find the regressions.

Gemini Enterprise addresses the first problem directly and gives you building blocks for the other two. This post builds on the platform view in part 1.

The standard to design for: every answer is drawn only from content the asking user is allowed to see, from the current version of that content, with a citation they can click — and when the content does not support an answer, the assistant says so.

The architecture

There are two retrieval paths and one access-control path. The platform path uses connectors that index into the Gemini Enterprise search index or federate to the source, answered by the built-in assistant and customizable RAG agent. The custom path is an ingestion pipeline into RAG Engine or Vector Search, read by an ADK agent on Agent Runtime that is registered into the same app. Underneath both: identities and groups from your identity provider, ACLs from each source, and a query-time filter on the person asking.

Knowledge-base architecture on Gemini Enterprise
Sources Connect Retrieve Agent Answer SharePointsites · OneDrive Confluence · Jiraspaces · issues Google WorkspaceDrive · Docs ServiceNowknowledge · records Cloud StoragePDFs · scans · exports Indexed connectors copy content + ACLs incremental sync Federated query at the source nothing copied Ingestion pipeline parse · chunk · embed Gemini Enterprise search index content · ACLs metadata RAG Engine corpus or Vector Search Gemini Enterprise app assistant built-in search customizable RAG agent Custom ADK agent on Agent Runtime registered in the app Cited answer sources linked as-of dates or refusal live query results Identity provider users · nested groups · guests ACL sync per-document principals Query-time filter on the asking user applied before anything reaches the model ACL path federated: the source enforces its own ACLs at query time
Two retrieval paths share one access-control path. Content reaches the model only after it has been filtered for the person asking; federated sources enforce their own permissions when queried.

Two points matter most. The custom path is additive: you register a specialist alongside built-in search, not instead of it. And the ACL path is not optional there: RAG Engine and Vector Search store what you give them, so if chunks do not carry their permitted principals, the custom agent becomes the hole in your permission model.

Components mapped to Google Cloud

LayerGoogle Cloud serviceWhat it does in the knowledge base
ExperienceGemini Enterprise appBuilt-in search and assistant, customizable RAG agent, Gemini Notebook Enterprise, Agent Gallery
ConnectPrebuilt connectorsSharePoint, Confluence, Jira, ServiceNow, Google Workspace and more; each source either indexed into the search index or federated at query time
Access controlPermissions-aware search, Cloud IAM, your identity providerAccess-controlled search results and generative answers; identities and groups resolved per user
Custom ingestionCloud Storage, Pub/Sub, Eventarc, Cloud Run, WorkflowsChange detection, parsing, chunking, metadata enrichment and upserts for content outside the connector path
Custom retrievalRAG Engine, Vector SearchManaged corpus and retrieval for a specialist agent; vector index with metadata filtering at larger scale
Custom agentADK on Agent Runtime, registered in the appDomain-specific retrieval, reranking, answer policy and refusal rules; admins choose which users can use it
Metadata & ownershipDataplex, BigQueryContent inventory, owners, effective and review dates, duplicate clusters, freshness metrics
ProtectionVPC Service Controls, CMEK with Cloud KMS, Model Armor, Sensitive Data ProtectionPerimeter around data and models, customer-managed keys, prompt-injection screening, PII discovery and masking
QualityAgent evaluation, online evaluation, Agent Simulation, ObservabilityGolden-set regression, sampled production scoring, synthetic multi-turn users, traces of every retrieval and answer

Indexed or federated: choosing per source

Gemini Enterprise connectors work in one of two ways. An indexed connector copies content — and the permissions attached to it — into Gemini Enterprise’s search index and keeps it in sync. A federated connector sends the query to the source system at question time and uses what comes back, without copying anything. Decide per source; most estates use both.

ConsiderationFavors indexedFavors federated
Answer qualityYou need ranking across sources, semantic retrieval over full text, and consistent snippetsThe source has strong native search for its own content type
Data residency and copyingA governed copy inside your Google Cloud boundary is acceptablePolicy or contract forbids copying the content out of the source
FreshnessMinutes-level lag from sync is acceptableContent changes by the minute and must be exact (ticket status, live records)
Volume and churnLarge, relatively stable corpora: policies, manuals, wikisEnormous or very high-churn data where a full copy is costly to maintain
PermissionsSource ACLs can be read and synced reliablySource permissions are complex or computed, and only the source can evaluate them correctly

The common pattern: index document estates (SharePoint, Confluence, Drive, knowledge articles) and federate transactional systems where the question is about a live record. Federated results are only as good as the source’s own search, so test them against your golden set; if it is weak and copying is allowed, index instead.

A useful default: index content people read, federate records people look up. Revisit any source whose federated answers score badly in evaluation.

Permission-aware retrieval and ACL sync

Permissions-aware access means results and generative answers draw only on what the asking user could open in the source. The guarantee holds while three things stay true: indexed ACLs match the source, source identities map to the identity that signs in, and groups resolve correctly. Each can break.

What breaks

  • Stale ACLs. Someone is removed from a confidential site at 9:00. If permissions are synced on a schedule, there is a window in which the index still lists them. Sync permission changes faster than content; for the most sensitive sources consider federation or exclusion.
  • Group resolution. Most enterprise ACLs are on groups, often nested, sometimes defined in the source system rather than the directory (SharePoint site groups, Confluence space permissions, Jira project roles). If a source-local group cannot be expanded, documents become invisible to the right people or mapped too broadly. Inventory where groups live before connecting.
  • Identity mapping. jsmith@corp.com in the directory is john.smith in an old wiki. Without a reliable mapping, documents silently disappear for the people who own them.
  • External and guest users. Contractors and partner guests appear in source ACLs with identities your workforce directory may not know. Decide explicitly whether guests use the assistant, and test documents shared only with external accounts.
  • “Everyone” links. Years of “anyone in the organization with the link” sharing mean the assistant will find documents nobody realized were broadly readable. The assistant makes old oversharing discoverable; review high-risk sites before connecting them.

The custom path carries its own ACLs

For content you ingest yourself into RAG Engine or Vector Search, store the allowed principals (users and expanded groups) as metadata on every chunk and filter on them inside the retrieval call, using the end user’s identity passed through from the Gemini Enterprise app — never the agent’s service identity. Filter before ranking: a forbidden document must never enter the model’s context. The agent harness in part 3 covers how Agent Identity and OAuth authorizations carry the user through.

Test permissions like a security control, because it is one. Keep “canary” documents restricted to test users and query them daily as allowed and disallowed users. A leaked canary is a severity-one incident.

Ingestion: parsing, chunking and metadata

For sources on the connector path, Gemini Enterprise handles extraction. For everything on the custom path — scanned archives, export dumps, documents with complex layout, content from systems without a connector — ingestion quality decides answer quality.

The ingestion pipeline for the custom path
Detect · fetch · parse Change eventPub/Sub · Eventarc Fetch & hashskip unchanged · dedupe Layout-aware parseheadings · tables · order OCR for scanswith confidence score Chunk · enrich · load Chunk by sectiontables kept whole Metadataowner · ACL · dates · type Sensitive-data scanmask or block Embed & upsertRAG Engine · Vector Search Quarantinehuman review queue low OCR confidence · blocked data fast lane: deletions & ACL changes
Every stage writes what it did to the document’s record, so a bad answer can be traced back to the parse, the chunk or the metadata that caused it. Deletions and permission changes skip the slow path.

Parsing: layout, tables and scans

Plain text extraction flattens a document: two-column layouts interleave, headers and footers repeat into every paragraph, and tables become a stream of numbers with no column meaning. Use a layout-aware parser that preserves heading hierarchy, reading order and table structure. Convert tables to a structured form (Markdown or HTML rows with header cells repeated) so that “the deductible for tier 3” can be found and quoted with its row and column. For scanned PDFs, run OCR, keep the per-page confidence, and route low-confidence pages to review rather than indexing noise. Record the parser version so improvements can be re-applied selectively.

Chunking

Fixed-size chunking is the most common source of “the answer was in the document but the assistant missed it.” Chunk on the document’s own structure: section and sub-section boundaries, with the heading path (“Travel Policy › International › Per diem”) prepended to each chunk so it carries its context. Keep tables, numbered procedures and definitions intact even if they exceed your target size. Keep a pointer to the parent section so the agent can expand context when a chunk is ambiguous.

Metadata that earns its keep

Every chunk carries source URL, document ID and version, heading path, owner, document type, effective and review dates, jurisdiction, allowed principals, parser version and content hash. Metadata is what lets you filter, boost the policy over a slide quoting it, explain (“effective 1 March”) and retire.

Built-in search or a custom RAG agent

Built-in search and the customizable RAG agent are the right first answer for most questions. Add a custom ADK agent on RAG Engine or Vector Search, hosted on Agent Runtime, when one of these signals appears — and keep the built-in path for everything else.

SignalBuilt-in is enoughAdd a custom RAG agent
SourcesCovered by prebuilt connectorsScanned archives, exports or systems with no connector
DocumentsMostly text pages and office documentsComplex tables, forms or layouts that need specialist parsing
Answer policyGeneral cited answers are acceptableStrict rules: answer only from the current effective version, always quote the clause, refuse outside scope
RetrievalDefault ranking performs well on the golden setDomain filters, reranking or multi-step retrieval are needed to pass evaluation
WorkflowQuestion in, answer outThe answer feeds a step: a draft reply, a checklist, a referral to a human
AuditStandard logging is sufficientRegulators need the exact chunks, versions and rules behind every answer

The custom agent is registered into the Gemini Enterprise app (Agents > Add agent, or the REST API), and admins choose which users can use it. Employees still start in one place: a general assistant, plus a few specialists where the answer policy justifies one.

Grounding, citations and refusing when unsupported

A grounded answer is one where every claim can be traced to a retrieved passage the user can see. Three rules make that real rather than aspirational.

1

Cite at the claim, not at the bottom

Citations attached to individual sentences, linking to the source document and, where possible, the section, let a reader verify the one sentence they care about. Show the document’s effective or last-modified date next to the citation.

2

Refuse when retrieval does not support an answer

If the retrieved passages do not answer the question, the correct response is “I could not find this in the sources available to you,” plus where to go next (the policy owner, an HR case, a team channel). In a custom agent, enforce this with a support check: before returning, verify that each claim is entailed by a cited passage and drop or flag what is not.

3

Do not let the model fill gaps from general knowledge

The model knows what a typical parental-leave policy says, not yours. Instruct and evaluate for sources-only behavior, and when sources conflict, show both citations rather than choosing silently.

Freshness SLOs and content lifecycle

A knowledge base has two freshness numbers, and both need a target, a measurement and an alert. Content freshness is the time from a change in the source to that change being retrievable. Permission freshness is the time from an access change in the source or directory to the index reflecting it. For example, a team might target a few hours for wikis, an hour for policies, and tighter for permissions. Federated sources are fresh by construction.

Re-index strategy

  • Incremental by default. Use change events or modified-since crawls; skip unchanged files by content hash.
  • Fast lane for deletions and ACL changes. A deleted or restricted document must leave the index faster than a new document enters it.
  • Targeted re-processing. When the parser, chunker or embedding model changes, rebuild into a new index version, evaluate it against the golden set, and switch over only when it is at least as good — blue-green for indexes.
  • Reconcile and measure. Compare index and source inventories on a schedule to catch missed events, and publish freshness percentiles per source in Cloud Monitoring.

Ownership, staleness and duplicates

A fresh pipeline does not help if the content is stale. Every document needs an owner and a review date. Build a content inventory in BigQuery from captured metadata and send owners three weekly reports: past review date, no active owner, and duplicate clusters. Owners confirm or retire; superseded versions are excluded; one canonical version wins. Documents often cited in answers marked wrong go first.

Retirement is a feature. Excluding an obsolete policy from retrieval improves answers more reliably than any prompt change.

Five use cases, and what each one demands

1. Employee policy and HR helpdesk

Leave, benefits, travel and expenses. The hard parts are jurisdiction and versions: Ontario differs from Texas, and the policy changed in January. Tag policies by country, employee type and effective date, filter on the employee’s attributes, and refuse personal-case questions (“how much leave do I have left?”) unless a governed tool answers them from the HR system.

2. Engineering knowledge

Runbooks, decision records, design docs, Jira and Confluence. Engineers expect the current runbook, not the one from two migrations ago. Index wikis, federate live ticket status, boost documents linked from the service catalog, and pair the assistant with the real-time context in part 4 when a question is about what is happening now rather than how things work.

3. Customer-service agent assist

The assistant suggests cited answers to a human agent from knowledge articles and product documentation. Measure acceptance rate alongside groundedness, and feed rejected suggestions into content review. The same knowledge base later powers the resolution agents described in part 5.

4. Sales enablement

Product sheets, pricing guidance, competitive notes and security questionnaires. The risks are stale pricing and unapproved claims. Restrict customer-facing drafts to approved collections, show effective dates, and keep internal competitive notes behind group ACLs.

5. Regulated knowledge: insurance underwriting guidelines

Underwriting guidelines are versioned by line of business, state and effective date, full of tables (rating factors, eligibility limits, referral thresholds) and frequently amended by bulletin. This is where a custom ADK agent earns its place: specialist table-aware parsing, strict filtering to the guideline in force for the risk’s state and effective date, answers that quote the clause and its version, and refusal plus referral to a senior underwriter when the guideline is silent. Every answer is logged with exact chunks and versions for audit.

Walkthrough: one question, end to end

An underwriter in a commercial lines team asks the underwriting guidelines agent in the Gemini Enterprise app: “For a restaurant risk in Florida, what is the maximum TIV we can write without referral?”

1

Identity and routing

The underwriter is signed in to Gemini Enterprise. The guidelines agent is registered in the app and enabled for the underwriting group only. The request reaches the ADK agent on Agent Runtime carrying the user’s identity.

2

Screening and filtered retrieval

Model Armor screens the prompt. The agent extracts the facets (line: commercial property, class: restaurant, state: Florida, date: today) and queries the RAG Engine corpus with metadata filters for document type = guideline, state in {FL, all}, effective date ≤ today and not superseded, and allowed principals containing the user or their groups. Top passages include the Florida table of referral thresholds and a bulletin amending it last month.

3

Screening retrieved content

Retrieved passages are screened by Model Armor for embedded instructions before they enter the model’s context.

4

Grounded answer with a conflict surfaced

The agent answers with the amended threshold from the bulletin, cites the bulletin clause and its effective date, notes that it supersedes the figure in the base guideline table (also cited), and states that risks above the threshold require referral. A support check confirms each sentence is backed by a cited passage.

5

Record and learn

The trace — query, filters, chunk IDs and versions, screening results, answer — goes to Cloud Logging. Online evaluation samples the turn and scores groundedness. If the underwriter marks the answer unhelpful, it enters the triage queue.

Security, identity and governance

  • VPC Service Controls. Put the corpus, indexes, ingestion buckets and agent runtime inside a service perimeter to block exfiltration even with valid credentials.
  • CMEK. Customer-managed keys in Cloud KMS for buckets, indexes and logs where policy requires it.
  • Model Armor against prompt injection inside documents. In a knowledge base, the most likely injection does not come from the user; it comes from a document. A wiki page saying “ignore your instructions” is retrieved like any other content. Screen retrieved passages as well as prompts, treat retrieved text as data in the agent’s instructions, and do not give knowledge agents tools with side effects they do not need.
  • Sensitive Data Protection (DLP). Scan for personal and regulated data at ingestion and mask or exclude it; run discovery on connector sources too, because an assistant makes forgotten spreadsheets findable.
  • Least privilege for agents. Each agent has its own identity with read-only access to its corpus; retrieval filters use the end user’s identity. Agent Registry and Agent Gateway, covered in part 3, keep the inventory and the policy enforcement point in one place.
  • Audit. Log query, user, filters, retrieved chunk IDs and versions, and the answer. Restrict who can read logs containing question text.

Evaluation: making quality measurable

A knowledge base is measurable when you can answer two questions with numbers: did we retrieve the right passages, and did we answer correctly from them? Separating the two matters, because the fixes are different — a retrieval miss is fixed in content, parsing, chunking or metadata; an answer error is fixed in instructions, answer policy or model choice.

The evaluation loop
Golden question set expected sources & answers must-refuse questions permission canaries Offline evaluation retrieval: recall@k · first-hit rank answers: correctness · grounded citation accuracy · right refusals Release gate no regression on any slice canaries: zero leaks Production online evaluation, sampled traces · user feedback freshness & refusal rates Triage by cause retrieval miss · stale or duplicate ACL error · parse error generation error Fix where it broke content owner · pipeline metadata · agent instructions add failure to golden set Agent Simulation promote grow
Every production failure is classified by cause, fixed where it broke and turned into a golden question, so the release gate gets stricter as the knowledge base matures.

The golden question set

Collect a few hundred real questions per domain from tickets, search logs and experts. For each, record expected sources, the key facts the answer must contain, and the persona asking. Include three special slices: must-refuse questions the sources do not answer, conflict questions where an old and a new version both exist, and permission canaries asked as users who must not see the answer. Content owners sign off on reference answers.

Metrics that separate retrieval from answering

MetricWhat it tells you
Retrieval recall@kWas at least one expected source in the top k passages given to the model?
Rank of first correct sourceHow high the right passage ranks; drives answer quality and context cost
Answer correctnessDoes the answer contain the reference facts, judged by an autorater calibrated against human grades?
GroundednessIs every claim supported by a retrieved passage?
Citation accuracyDoes each citation actually support the sentence it is attached to?
Correct refusal rateOn must-refuse questions, did it decline? On answerable ones, did it avoid refusing?
Permission leakageCanary content in answers for disallowed users; the target is zero
FreshnessSource-change-to-retrievable percentiles per source, against SLO

Offline, simulated and online

Run the golden set through agent evaluation, with multi-turn autoraters, on every change to connectors, parsing, chunking, prompts or models; block any regression. Agent Simulation adds synthetic users who follow up, rephrase or push back on a refusal — failures that appear on turn three. In production, online evaluation scores sampled turns and thumbs-down feedback goes to triage. Calibrate autoraters against human grades before trusting them.

A 90-day rollout plan

30

Days 1–30: one domain, measured

Pick one domain with clear owners and real demand — HR policy is a common choice. Inventory its sources, owners and groups; run an oversharing review; decide indexed or federated per source; connect them. Build the golden set with the domain team, including canaries and must-refuse questions. Baseline the built-in assistant against it. Set content and permission freshness SLOs and start measuring.

60

Days 31–60: close the gaps

Fix what the baseline exposed: retire duplicates, assign missing owners, add metadata, tune the customizable RAG agent. If evaluation shows the built-in path cannot meet the bar for a sub-domain, build the custom ingestion pipeline and an ADK agent for it, with ACL metadata and Model Armor screening from day one. Wire up VPC Service Controls, CMEK and DLP scanning. Launch to a pilot group with feedback capture and online evaluation on.

90

Days 61–90: operate and expand

Put the release gate into the deployment pipeline. Start weekly content-lifecycle reports to owners. Review production failures by cause and grow the golden set. Go live for the first domain and start the second with the same playbook, reusing the pipeline, evaluation harness and controls.

Failure modes we plan for

1. Connecting everything on day one. Thousands of sites with unknown owners and years of oversharing become instantly searchable. Connect by domain, after an oversharing review, with owners named.

2. A custom agent that ignores ACLs. The pipeline loads documents into a corpus, the agent queries it with its own service identity, and every user can now read everything in it. Chunks carry principals; retrieval filters on the end user.

3. Deletions that never propagate. A recalled document stays retrievable because the pipeline only handles creates and updates. Deletions and ACL changes take the fast lane, and reconciliation catches what events miss.

4. Tables flattened into noise. Rate tables and eligibility grids lose their structure, so the assistant quotes the wrong row with confidence. Layout-aware parsing, tables kept whole, and table questions in the golden set.

5. Three versions of the truth. Old and new policies both rank, and the assistant picks one. Effective dates, supersession metadata and canonical versions; surface conflicts instead of choosing silently.

6. Instructions hidden in documents. A retrieved page tells the model to do something else. Screen retrieved content with Model Armor, treat it as data, and keep knowledge agents free of unneeded tools.

Readiness checklist

AreaReady when
ScopeFirst domain chosen, with named content owners and a business sponsor
ConnectorsEach source assigned indexed or federated, with the reason recorded
PermissionsGroup sources inventoried, identity mapping tested, guests decided, oversharing reviewed
IngestionLayout-aware parsing, table handling, OCR confidence and quarantine in place for the custom path
MetadataOwner, effective and review dates, type, ACL principals and version on every chunk
FreshnessContent and permission SLOs defined per source, measured and alerted on
LifecycleWeekly stale, ownerless and duplicate reports reach owners; supersession is honored in the index
Answer policyClaim-level citations, refusal when unsupported, conflicts surfaced
SecurityVPC Service Controls perimeter, CMEK where required, Model Armor on prompts and retrieved content, DLP scanning
EvaluationGolden set with must-refuse, conflict and canary slices; release gate; online evaluation and Agent Simulation in use
OperationsTraces for every turn; failures triaged by cause and added to the golden set

The bottom line

Gemini Enterprise removes much of the plumbing that used to consume knowledge-base projects: connectors, a permissions-aware index, cited answers and an app employees already use, with a clean way to add specialist ADK agents on RAG Engine or Vector Search when a domain needs one. What the platform cannot do for you is decide which sources to index or federate, keep your groups and owners in order, retire the policy that was replaced last spring, or tell you whether this week’s change made answers better. Treat permissions as a security control with canaries, freshness as an SLO, content as something with owners and an end date, and quality as a number with a release gate. That trust is what lets the knowledge base become the foundation for the agents in the rest of this series.

At VerticalServe we have spent seven years building data and AI platforms on Google Cloud and run production agentic systems in insurance. Our InsightAsk and InsightRAG accelerators package the patterns described here and run alongside Gemini Enterprise, not in place of it.

Building a knowledge base people can trust?

We help teams stand up permission-aware, measurable knowledge bases on Gemini Enterprise — connectors, ingestion, custom RAG agents and the evaluation loop — using our InsightAsk and InsightRAG accelerators where they save time.

Data & AI on Google Cloud
Back to blog