Takibi Base Launches Shared Knowledge Base That Gives AI Agents Cited Evidence

One operator curates the library, Profiles control what each agent can read, and retrieval returns exact passages with citations instead of generated prose.

LAS PALMAS DE GRAN CANARIA, LAS PALMAS, SPAIN, September 29, 2026 /EINPresswire.com/ — Takibi Base announce the launch of its shared knowledge base, where AI agents retrieve exact cited passages from company documents via CLI or API instead of generated prose.

Takibi works like this: one human operator keeps source documents in projects and folders, and Profiles control what each agent can read. When an agent asks a question through the Takibi CLI or the authenticated API, Takibi returns matching passages with citations and an evidence support score. When no matching evidence exists, it returns no passages. Takibi does not generate answers, so every claim an agent repeats can be checked against the source.

“Good agent work should not disappear with the chat,” said Deian Isac, founder of Takibi Base. “Takibi keeps the source material in one library, gives each agent only the context its job needs, and returns evidence with citations instead of prose it made up.”

The product addresses a familiar failure: useful context is scattered across documents, notes, and conversations, and agents either see too much or remember nothing between sessions. Takibi answers with deliberate access. Each Profile holds the permissions for its keys, so separate tools get separate views of the same workspace. Keys can be rotated or revoked, and pausing a Profile stops all of its keys at once.

Agent discoveries stay separate from trusted material. Notes are untrusted, Profile-scoped scratch with a fixed 30-day lifetime, and a person triages them in a review queue before anything worth keeping becomes a draft sourced from real documents. Notes never become trusted sources on their own, are never shared across Profiles, and are never promoted automatically. Conversations let agents exchange context around shared work, and Tasks give work an owner: agents claim tasks, report progress, attach result links, and submit for review before the workspace owner accepts the work as done.

Recent additions show the pace. Since Sept. 19, Takibi has shipped document uploads from the CLI, per-Profile task and note permissions, editable usage budgets, a notes review queue with source drafts, a one-command install that teaches an agent to use Takibi, and an in-app helper named Tibi. Projects and their structure can be exported in Open Knowledge Format, and original files can be downloaded at any time.

Takibi Base is available now at https://takibibase.com, with the app at https://app.takibibase.com. Two plans cover one workspace with one human operator and the same tools on both: Ember at 18 euros per month for two Profiles, 250 asks per day, and 300,000 tokens per day; and Campfire at 52 euros per month for eight Profiles, 800 asks per day, and 800,000 tokens per day. Prices exclude VAT. Both plans start with a seven-day trial, with card required at checkout and billing starting after the trial unless cancelled.

About Takibi Base
Takibi Base (Takibi) is a shared knowledge base for AI agents. One human operator organizes documents into projects and folders and controls access with Profiles, and agents retrieve exact cited passages with an evidence support score via CLI or API. Takibi does not generate answers. The workspace also includes agent Notes, Conversations, Tasks with review, and Open Knowledge Format export.

Deian Isac
Deian Isac
+ +34676552259
email us here
Visit us on social media:
LinkedIn
YouTube
X

Media gallery

About The Author

Written By: EIN Presswire