Your AI coding agent is missing context.

DevsContext fetches from Jira, meeting transcripts, and your docs, then synthesizes it into one context block your AI actually understands.

Before vs After

Without DevsContext

You: "Work on PROJ-123"

Claude: Opens Jira MCP, sees ticket title. Guesses the implementation. Doesn't know about last week's architecture decision. Uses wrong error handling pattern. You spend 30 minutes correcting.

With DevsContext

You: "Work on PROJ-123"

Claude: Gets requirements, acceptance criteria, the decision from sprint planning about which queue to use, the architecture doc showing where to add code, and your team's error handling standards. Writes it right the first time.

Connect Your Tools

DevsContext pulls context from where your team works

๐Ÿ“‹

Jira

Tickets, comments, linked issues

๐ŸŽ™๏ธ

Fireflies

Meeting transcripts, decisions

New
๐Ÿ’ฌ

Slack

Team discussions, threads

New
๐Ÿ“ง

Gmail

Email threads, decisions

๐Ÿ“

Local Docs

Architecture, standards, ADRs

Soon
๐Ÿ“

Notion

Documentation, wikis

Soon
๐Ÿ“Š

Linear

Issue tracking

Soon
๐Ÿ“–

Confluence

Team documentation

What Claude Code sees

One synthesized context block with everything it needs

## Task: PROJ-123 โ€” Add retry logic to payment webhook handler

### Requirements
1. Implement exponential backoff for failed webhook deliveries
2. Max 5 retry attempts over 24 hours
3. Dead-letter queue for permanently failed webhooks

Acceptance criteria: [Jira PROJ-123]
- [ ] Webhooks retry with backoff (1min, 5min, 30min, 2hr, 12hr)
- [ ] Failed webhooks move to DLQ after 5 attempts
- [ ] Dashboard shows retry metrics

### Key Decisions
Use SQS with visibility timeout for retry scheduling, not cron.
Decided by @sarah in March 15 sprint planning. Rationale: SQS handles
timing natively, reduces operational overhead. [Meeting: Sprint 23 Planning]

### Architecture Context
Webhook flow: PaymentController โ†’ WebhookService.dispatch() โ†’ SQS queue
โ†’ WebhookWorker.process() โ†’ external endpoint.

Add retry logic in WebhookWorker.process() at:
src/workers/webhook_worker.ts:45-80 [Architecture: payments-service.md]

### Coding Standards
- Use Result<T, WebhookError> pattern, don't throw exceptions
- Retry delays: use calculateBackoff() helper from src/utils/retry.ts
- Tests: mock SQS with @aws-sdk/client-sqs-mock [Standards: typescript.md]

### Team Discussions
#payments-team (2024-03-18)
@sarah: "For the webhook retries, let's use SQS visibility timeout
rather than a separate retry table. Simpler and handles timing natively."
@mike: "Agreed. I'll update the design doc." [Slack: #payments-team thread]

Built for Teams

Features that make AI coding assistants actually useful

Pre-processing Agent

Build rich context proactively. When tickets move to "Ready for Dev", context is already waiting.

devscontext agent start
# Polls Jira, pre-builds context

Plugin System

Add custom sources or synthesis strategies. Publish as pip packages with entry points.

# pip install devscontext-linear
# Auto-discovered at startup

Quick Start

pip install devscontext

devscontext init
# Configure Jira, Fireflies, local docs

claude mcp add devscontext -- devscontext serve
# Connect to Claude Code

# Then in Claude Code:
> work on PROJ-123
View on GitHub

Open source. MIT licensed. Built for developers.