Inspiration
Research teams already coordinate experiments, lab meeting notes, protocol questions, and analysis decisions in Slack. The problem is that important action items and QC risks get buried in long message threads. LabSignal was built to make Slack itself a lightweight research operations surface.
What It Does
LabSignal is a Slack-native agent for neuroscience teams. It can:
- Build structured lab handoffs from messy updates
- Extract owners, tasks, and deadlines
- Flag QC, schedule, governance, and reproducibility risks
- Generate experiment checklists
- Search a small neuroscience protocol knowledge base
- Expose the same capabilities through an MCP-compatible tool server
Example: a researcher can run /labsignal brief ... and get a summary, action items, risk flags, and relevant protocols in one Slack response.
How We Built It
LabSignal uses Slack Bolt for Python with Socket Mode, so it can run locally without a public webhook during development. The agent routes Slack slash commands and app mentions to deterministic local tools. Those same tools are exposed through an MCP server, including search_protocols, extract_action_items, summarize_update, detect_risks, build_research_brief, and plan_experiment.
Challenges
The main challenge was making the agent feel useful inside Slack rather than like a generic chatbot. I focused on clear, structured responses that match real research workflows: lab handoffs, QC follow-up, SOP lookup, and experiment planning.
Accomplishments
I built and tested a working Slack app in a Slack developer sandbox, added slash command and mention support, implemented MCP-compatible tools, created a clean architecture diagram, and recorded a working demo.
What's Next
LabSignal could connect to electronic lab notebooks, GitHub issues, Notion/Google Drive protocols, and real lab data QC pipelines. It could also add persistent memory so it remembers project context across weeks of research activity.
Log in or sign up for Devpost to join the conversation.