Guru in Slack: Find Team Knowledge and Keep It Current

Guru is a knowledge management tool that lets teams save verified information as cards and surface them inside Slack. When a teammate asks a question in Slack, a Guru card can answer it without requiring them to open another browser tab or search through a wiki. Tools like fenbs hold the project decisions and current context that should be available to AI assistants but belong outside Guru's scope — active work state, rules for this project, and Where we left off.

The combination is useful: Guru handles stable reference knowledge; a connected AI assistant handles active work context. The challenge is keeping both current, because outdated knowledge in either place produces confidently wrong answers.

What Guru in Slack does

The Guru Slack integration gives your team two main capabilities: the Guru bot can answer questions asked directly in Slack by searching your knowledge base, and users can save a Slack message as a Guru card directly from the message menu.

Guru also suggests cards based on the questions being asked in a channel. When it detects that someone is looking for information that a Guru card covers, it can suggest the card automatically. This works best in channels with a clear topic — onboarding, customer support, engineering FAQs — rather than general team channels.

Check Guru's Slack integration documentation↗ (opens in a new tab) for the current setup steps and feature list. The integration has changed between versions and what is described in older guides may not match the current behavior.

How to use Guru cards inside Slack

1. Install the Guru Slack app from your Guru workspace settings. You need admin access to both Guru and Slack.
2. In Slack, use the /guru command or mention @Guru in a message to search your knowledge base. Guru returns matching cards inline.
3. To save a Slack message as a Guru card, hover over the message and use the "Add to Guru" option in the message actions menu.
4. When Guru suggests a card in a channel, click it to expand the card without leaving Slack.
5. Share a Guru card link directly in Slack; teammates can open it inline.

Who owns each knowledge card

Every Guru card has an owner — the person responsible for keeping it accurate. When a card's information changes, Guru sends the owner a verification reminder. Cards that are not verified within a set interval are flagged as potentially outdated.

Card ownership is one of the most important things to set up correctly. Without an owner, cards drift out of date and nobody knows whose job it is to fix them. Assign ownership when you create the card, and use Guru's groups or collections feature to assign ownership by team or topic so no single person owns too many cards.

Keeping cards current

Outdated cards are more harmful than no cards. A team that trusts Guru will act on what it says without double-checking. An incorrect card propagates the wrong answer confidently and at scale.

The most reliable update trigger is a process change: when a policy, a tool, or a procedure changes, review the relevant cards immediately rather than waiting for the verification reminder. Build that review into your offboarding process too — when someone leaves, audit the cards they owned.

For knowledge that changes frequently — project status, active decisions, current tasks — a Guru card is the wrong format. Those things belong somewhere they are expected to change, not in a stable reference card that will drift. See workflow automation tools for tools that handle dynamic state better.

What Guru and Slack don't hold for AI context

Guru is a reference library. It holds stable, verified facts. It is not the place to track where the team left off on a project, which decisions apply to the current sprint, or which tasks are in progress. Those belong in an active work record.

When an AI assistant is connected to Slack or Guru, it can answer questions from the knowledge base well. It cannot tell the team what was decided in Tuesday's meeting, which tasks are blocked, or what the next step on the current project is — unless that information lives somewhere it can read.

fenbs keeps the active work record: projects, tasks, Decisions and rules, lessons, and Where we left off. That record is meant to change as the work changes, unlike Guru's stable reference cards. Both tools have a role — Guru for what stays true, fenbs for what is true right now. See agentic workflows for how these tools fit into a broader automated workflow.

How to keep project decisions with fenbs

1. Create a free account and add a project for the current team work.
2. Write the active decisions — the ones that affect the next two weeks of work — as Decisions and rules. Do not duplicate the stable reference information already in Guru.
3. Add current tasks and their status to the project.
4. Connect your AI assistant over MCP from Integrations. Ask it to read the project context before helping your team with active work.
5. Use Where we left off to record the project's current state after significant sessions so the next one starts with a clear picture.

fenbs does not replace Guru. Use Guru for the team's stable reference knowledge and fenbs for the active work context. Point your AI assistant to the right source for each type of question: Guru for "how do we handle customer refunds?" and fenbs for "where are we on the refund workflow project?"