# Peter Wilson & Davide Eynard - cq - Stack Overflow for Agents - AI Native DevCon June 2026

## Executive summary

The session introduces CQ (Mozilla.ai's proposal), a system designed as a 'Stack Overflow for agents.' Its core purpose is to standardize and share knowledge units (KUs) across autonomous AI agents—locally, within an organization, or publicly. This prevents agents from repeating mistakes, wasting tokens, and ensures that lessons learned by one agent can benefit all others, thereby improving the reliability of complex automated workflows.

## Key takeaways

- Knowledge Unit (KU) Standardization: A KU is a standardized knowledge artifact (stored in JSON format) capturing solutions to novel problems. It includes domains, insights, actions taken, summaries, and metadata (languages/frameworks).
- Agentic Context Management: Effective agent performance relies heavily on context injection. The system aims to move beyond simple memory files by allowing agents to query a centralized knowledge base for relevant solutions before starting a task.
- Layered Sharing Model: CQ supports three levels of knowledge sharing: local (SQLite database, no review), private/team (requires user authentication and human-in-the-loop review), and public commons (CQ Exchange).

## Technical details

- Knowledge Unit Schema: KUs are structured to capture problem context, including domains, insights, actions taken, summary, and metadata regarding languages and frameworks. This standardized format is key for cross-agent knowledge transfer.
- CQ Workflow (Remote Mode): When an agent encounters a novel issue, it proposes a KU to the CQ server. The proposal must undergo Human-in-the-Loop review before becoming available for querying by other agents via a dedicated skill plugin.
- Security and Mitigation: To mitigate risks (e.g., PII leakage, identity spoofing), proposed defenses include OAuth stride, cryptographic signing of KUs, layered guardrails pipelines, and mandatory human review.

## Practical implications

- For build engineering, CQ provides a mechanism to formalize and share debugging knowledge. Instead of relying on tribal knowledge or manual documentation, agents can automatically propose solutions (KUs) when they solve complex integration issues, improving CI/CD reliability.
- It helps reduce the time spent troubleshooting by allowing new automation runs to query past successful fixes for similar domain problems.

## Topics

AI Agents, Knowledge Graph, Open Source AI, LLM Orchestration, DevOps, CQ Exchange, Mozilla AI Open Source CQ Repo

Source: https://www.youtube.com/watch?v=AHIY1XccX_E