recon
0

A PM just asked a question. An engineer just lost 2 hours.

Stop asking engineers.
Ask Recon.

Get to the bottom of anything. Recon checks your database, code, tickets, and docs, then delivers the answer with full evidence.

For PMs, CS leads, support, and ops at SaaS companies.

Watch a 30-minute task done in 30 seconds

From hours of back-and-forth to instant answers.

The cost of the status quo

A customer files a ticket. The CS lead can't answer it because the real answer isn't in the knowledge base (it's never up to date). It's in the database, the code, or a ticket engineering is already tracking. So they ping an engineer. The engineer drops what they're doing, spends 30 minutes digging across four systems, and pastes the answer. Multiply that across a growing customer base and it becomes the bottleneck that slows down everyone.

How it works

From setup to first investigation in under 5 minutes.

01

Add Recon to Slack

One click to install. Recon joins your workspace and is ready to answer questions in any channel — no terminal, no config.

Add to Slack
recon is ready in your workspace
02

Connect your tools

Link your Postgres database, GitHub repos, Notion docs, Linear tickets, and more. Read access to your data, set up in 5 minutes.

PostgreSQL
GitHub
Linear
Sentry
HubSpot
Notion
03

@mention Recon, get answers

Ask anything in plain English. Recon queries your database, reads your code, checks tickets and docs, and replies with a full root cause analysis.

1
SQL Query
2
Code Read
3
Ticket Check
4
Root Cause

What will your team investigate?

Customer Success
Which of our enterprise accounts haven't logged in this week?
1
Queried users table for last_active timestamp
2
Filtered to enterprise plan accounts
3
Cross-referenced with open ticket history
Result

6 accounts inactive 7+ days · $48,000 ARR at risk · 2 have open support tickets

Product demos

Real investigations, start to finish.

recon
askrecon.com
We're launching ListN for Teams — check ticket status, are there real users, how does assignment work, and draft a launch announcement
1
Searched Linear (6 queries) → 9 tickets · 8 Done · CHA-150 (role management UI) in Backlog
2
Read Notion "ListN for Teams — Product Overview" (209 lines)
3
Queried DB → 1 org (UseListN) · 2 members · 7 tasks in production
4
Read capture-orchestrator.ts + task-manager.ts → traced full assignment flow
epic-1_launch_announcement.mdready to download
3.0K · Linear + Notion + DB + codebase · full launch brief

Ask anything about a feature. Recon checks tickets, reads your docs, queries the DB, traces the code — then drafts the brief.

Everything your team
needs to investigate

Connect once. Then anyone can ask questions in plain English.

Natural language queries

Ask like you'd ask a senior engineer

“Which customers churned last month and what were their top support complaints?” Recon understands context, runs queries, reads code, and connects the dots.

How does our assignment logic work?Why did this customer's invoice change?Who signed up this week and from where?
Full audit trail

See every step

Every SQL query run, every file read, every ticket checked — visible and auditable. No black boxes.

1
Queried subscriptions table
2
Cross-referenced 8 Intercom tickets
3
Found root cause in webhook handler

Connect any source

Postgres, GitHub, Notion, Linear, Jira, Intercom, Sentry, Stripe, HubSpot. Connect in minutes. Recon reads your data and can create tickets and docs when you ask.

Export anything

MBRs, churn summaries, account health pulls — Recon generates the file and downloads it directly from chat. No copy-pasting, no formatting.

Multiple workspaces

Separate your clients, teams, or environments. Each workspace has its own connections, conversations, and access controls.

Gets smarter over time

After each investigation, Recon saves what it learned about your schema and data. Next time it skips the discovery phase and goes straight to the answer.

Proactive monitors

Recon watches your accounts on a schedule and shows up in Slack with problems it found before anyone asked.

Works with your stack.

Connect once. Recon reads across all your sources simultaneously — code, data, tickets, docs.

PostgreSQL

SQL · Tables · Schema

GitHub

Code · Commits · PRs

Notion

Docs · Wikis

Intercom

Conversations · Contacts

Sentry

Errors · Monitors

Confluence

Pages · Spaces

Linear

Tickets · Cycles

Jira

Issues · Sprints

Supabase

Postgres · Auth

Stripe

Payments · Customers

HubSpot

Contacts · Deals

More integrations coming soon. Don't see what you need? Reach out.

Security-first, by design

Databases and code are strictly read-only. Write actions (creating tickets, posting to Slack) require explicit approval before they execute.

Read-only by default

Databases and code are strictly read-only. Write actions require your approval first.

AES-256 encryption

Credentials encrypted at rest, injected ephemerally

Full audit trail

Every query and file access logged and visible

Isolated sandboxes

Each investigation runs in its own Firecracker microVM

Bring your own key

Your API key, your LLM costs, your data

Defense in depth

MCP read-only flagsPostgreSQL read-only roleSystem prompt constraintsEphemeral sandboxesAES-256 encrypted credentials

Simple, transparent pricing

We set it up with you. Bring your own API key on paid plans.

Slack bot
Unlimited members
All integrations
Sandbox execution
Free
Free

No credit card required

  • 50 queries total
  • 1 workspace
  • 3 connections
  • Slack bot included
Starter
$19/ month

Bring your own API key

  • 50 queries/day
  • 1 workspace
  • 5 connections
  • Slack bot included
Popular
Pro
$49/ month

Bring your own API key

  • 150 queries/day
  • 3 workspaces
  • 15 connections
  • Proactive monitors
Business
$199/ month

Bring your own API key

  • 500 queries/day
  • 10 workspaces
  • 50 connections
  • Proactive monitors
  • Audit log & usage analytics

Every plan includes a guided setup call with our team.

Common questions

The things every team asks before connecting.

Your answer is one question away.

We set it up with you. One call, fully connected.

Most teams are live in under 30 minutes.

Curious? Ask an AI about Recon