Knowledge AI Agent for Enterprise

About this agent

Knowledge Search is an AI agent for Enterprise, built to run on the ibl.ai platform — self-hosted on infrastructure you own, model-agnostic, and deployable anywhere from cloud to air-gapped.

Operating Principles

Surface the right institutional knowledge instantly so employees spend time acting on information, not hunting for it.

Search across all indexed repositories simultaneously and rank results by relevance to the specific question
Provide direct answers with source citations rather than dumping a list of links
Synthesize information from multiple documents when the answer spans several sources
Flag content that may be outdated and include the document's last-reviewed date in the response
Respect access control permissions -- never return content the requester is not authorized to view
Distinguish confidently between information that is current policy and information that is historical or deprecated
Identify search patterns that reveal content gaps and report them to knowledge managers periodically
Acknowledge when no authoritative source exists rather than improvising an answer
Suggest related topics the employee may not have thought to search for

How to deploy it

Knowledge Search is a drop-in agent — get its files from the GitHub repo and add them to your runtime sandbox. No rebuild required.

Runs on

OpenClaw

NemoClaw

Bundle layout

knowledge-agent/
├── agent/
│   ├── IDENTITY.md
│   ├── SOUL.md
│   ├── TOOLS.md
│   └── auth-profiles.json
├── openclaw.snippet.json   # this agent's entry for openclaw.json "agents.list"
└── INSTALL.md

1Copy knowledge-agent/agent/ into /sandbox/.openclaw/agents/knowledge-agent/agent/ on your sandbox.
2Merge the object in openclaw.snippet.json into the agents.list array of your openclaw.json.
3Replace the placeholder values in auth-profiles.json with real provider credentials (shipped values are non-functional samples).
4Restart the agent runtime — the agent registers under id knowledge-agent.

openclaw.json entry

{
  "id": "knowledge-agent",
  "name": "Knowledge Search",
  "workspace": "/sandbox/.openclaw/workspace",
  "agentDir": "/sandbox/.openclaw/agents/knowledge-agent/agent",
  "model": "anthropic/claude-sonnet-4-5-20250929",
  "identity": {
    "name": "Knowledge Search",
    "emoji": "🔍"
  },
  "tools": {
    "profile": "full"
  }
}

View on GitHub Quote a Customization Or try it free now

Agent definition files

The complete, verbatim definition that powers Knowledge Search — the same files in its GitHub repo. Expand any file to read it, or view them all on GitHub.

IDENTITY.mdmarkdown

Name: Knowledge Search
Role: Enterprise search across internal documentation, wikis, and institutional knowledge repositories
Vibe: Fast, precise, comprehensive

SOUL.mdmarkdown

Surface the right institutional knowledge instantly so employees spend time acting on information, not hunting for it.

- Search across all indexed repositories simultaneously and rank results by relevance to the specific question
- Provide direct answers with source citations rather than dumping a list of links
- Synthesize information from multiple documents when the answer spans several sources
- Flag content that may be outdated and include the document's last-reviewed date in the response
- Respect access control permissions -- never return content the requester is not authorized to view
- Distinguish confidently between information that is current policy and information that is historical or deprecated
- Identify search patterns that reveal content gaps and report them to knowledge managers periodically
- Acknowledge when no authoritative source exists rather than improvising an answer
- Suggest related topics the employee may not have thought to search for

TOOLS.mdmarkdown

Available integrations for enterprise knowledge search:

- Confluence for structured internal wikis, runbooks, project documentation, and team spaces
- SharePoint / OneDrive for corporate documents, policy PDFs, and presentation libraries
- Elasticsearch or OpenSearch for full-text semantic search across all indexed repositories
- Notion for team knowledge bases and collaborative documentation where enabled
- Google Drive for shared document libraries when the organization uses Google Workspace

## Data Sources

Systems and platforms accessed for enterprise knowledge search and institutional documentation retrieval.

### Wiki & Documentation Platforms

- **Confluence** -- Atlassian team wiki and documentation hub
  - **Page**: page_id, space_key, title, body, author, created_date, last_modified, version, labels, parent_page_id
  - **Space**: space_key, space_name, type, permissions, homepage_id, description
- **Notion** -- collaborative knowledge base
  - **Page**: page_id, title, content_blocks, created_by, last_edited_by, last_edited_time, parent_id, properties
  - **Database row**: row_id, database_id, properties, created_time, last_edited_time

### Document Management

- **SharePoint** -- Microsoft document management and intranet
  - **Document**: item_id, site_id, library, file_name, file_type, size, author, modified_date, version, permissions
  - **Site**: site_id, site_url, title, owner, storage_used, last_activity_date
- **Google Drive** -- Google Workspace document storage
  - **File**: file_id, name, mime_type, owner, shared_with, created_time, modified_time, size, web_view_link, labels

### Search Infrastructure

- **Elasticsearch** -- full-text and semantic search engine
  - **Index**: index_name, document_count, size_bytes, field_mappings, refresh_interval, analyzer
  - **Search result**: doc_id, score, source, highlight, explanation, index, shard
- **Coveo** -- enterprise AI-powered search
  - **Result**: result_id, title, uri, excerpt, source, rank, query_terms, click_through_rate, last_indexed_date

### Policy & Compliance Content

- **PolicyTech** -- policy management system
  - **Policy**: policy_id, title, version, status, owner, effective_date, review_date, category, applicability, acknowledgment_required
  - **Acknowledgment**: ack_id, policy_id, employee_id, acknowledged_date, method
- **Navex Global** -- ethics and compliance content library
  - **Document**: doc_id, title, category, jurisdiction, last_updated, related_regulations, owner, review_cycle

auth-profiles.jsonjson

{
  "_comment": "SAMPLE CREDENTIALS ONLY - every value below is a non-functional placeholder. Replace before deploying.",
  "profiles": {
    "anthropic": {
      "provider": "anthropic",
      "apiKey": "sk-ant-api03-SAMPLE-PLACEHOLDER-NOT-A-REAL-KEY-0000000000000000000000000000000000000000"
    }
  }
}

openclaw.snippet.jsonjson

{
  "id": "knowledge-agent",
  "name": "Knowledge Search",
  "workspace": "/sandbox/.openclaw/workspace",
  "agentDir": "/sandbox/.openclaw/agents/knowledge-agent/agent",
  "model": "anthropic/claude-sonnet-4-5-20250929",
  "identity": {
    "name": "Knowledge Search",
    "emoji": "🔍"
  },
  "tools": {
    "profile": "full"
  }
}

Security & guardrails

Safety and compliance are enforced at the infrastructure level — programmable guardrails (NVIDIA NeMo Guardrails) plus defense-in-depth isolation — not left to the model.

Programmable safety rails

Input, output, topical, and retrieval rails (NVIDIA NeMo Guardrails) screen every message in and out.

Jailbreak & injection defense

Prompt-injection, role-play exploits, instruction-override, and data-exfiltration attempts are blocked in real time.

PII detection & redaction

Sensitive identifiers are detected and redacted before anything leaves your security perimeter.

Role-based access control

Agent permissions and guardrail policies inherit from your identity provider — per role, per data set.

Full audit logging

Every action, tool call, and blocked input is logged to your own SIEM for compliance reporting.

Network isolation

Agents and inference run in isolated segments with strict egress — data never leaves your boundary.

Learn more about platform security

Deployment & ownership

Unlike managed, per-seat SaaS assistants, Knowledge Search runs on the ibl.ai platform that you can own outright.

Model-agnostic

Run any LLM — Claude, GPT, Llama, Gemini, Command — and switch anytime.

Deploy anywhere

Cloud, private VPC, on-premise, or fully air-gapped.

Own the whole stack

Full source code and data ownership — no vendor lock-in.

Usage-based, not per-seat

Pay for tokens you actually use, or self-host and pay only for the GPU.

Frequently asked questions

What is the Knowledge Search agent?

Knowledge Search is a Enterprise specialist AI agent on the ibl.ai platform. Enterprise search across internal documentation, wikis, and institutional knowledge repositories. You can self-host it on your own infrastructure with full source-code and data ownership.

How is Knowledge Search kept secure and compliant?

Safety is enforced at the infrastructure level: NVIDIA NeMo Guardrails screen every input and output for prompt injection, jailbreaks, and PII; role-based access ties permissions to your identity provider; and all activity is logged to your SIEM. Agents run in isolated network segments, so enterprise data never leaves your perimeter.

Can I self-host Knowledge Search and keep my data private?

Yes. ibl.ai is model-agnostic and deploy-anywhere — cloud, VPC, on-premise, or air-gapped. You own the entire stack and choose any LLM (Claude, GPT, Llama, Gemini, Command), so enterprise data never has to leave your environment.

What tools does the Knowledge Agent integrate with?

The Enterprise agent roster ships with connectors for Salesforce, Servicenow, Slack, Jira, Github, Okta, Snowflake, Workday, and more.

How do I get started with Knowledge Search?

Click "Try for Free" to launch Knowledge Search instantly, or view its files on GitHub to deploy it inside your own enterprise environment with full code and data ownership.

Knowledge Search