How M6P works

Align coding agents through access to team context.

M6P keeps Skills and operational Knowledge private to your workspace, ready for agents to actively retrieve through your team’s MCP access while work is happening.

1. Connect an agent

Add M6P to the coding agent your team already uses.

2. Retrieve team context

Agents actively retrieve the approved Skills, policies, playbooks, and references that apply to their work.

3. Upload and keep the important parts

M6P extracts submitted content in memory and sends the extracted text to the Amazon Bedrock model configured for M6P’s AWS account. M6P does not retain the uploaded file or submitted raw text; it retains the source reference, bounded evidence excerpts, compiler provenance, and reviewable change set. See how Knowledge source data is handled.

Put the right knowledge within reach

When an agent needs help with a decision, practice, or constraint, it can actively retrieve the relevant approved Knowledge for the task at hand—instead of spending tokens on broad, traditional searches.

The result is a more consistent and efficient source of context for reviewing, building, releasing, and operating software together.

What your workspace can do

See connection readiness and factual retrieval Activity.
Maintain private team Skills and personal experiments.
Review evidence-backed Knowledge proposals before publishing them.

M6P is currently available by invitation during our pilot. We’re working closely with early teams to make the core experience reliable and useful.