Skip to content

feat: add Sift agent execution governance MCP server#3641

Open
walkojas-boop wants to merge 1 commit intomodelcontextprotocol:mainfrom
walkojas-boop:add-sift-governance-server
Open

feat: add Sift agent execution governance MCP server#3641
walkojas-boop wants to merge 1 commit intomodelcontextprotocol:mainfrom
walkojas-boop:add-sift-governance-server

Conversation

@walkojas-boop
Copy link

@walkojas-boop walkojas-boop commented Mar 19, 2026

Sift -- Agent Execution Governance

Sift is a cryptographic execution authorization layer for AI agents. This MCP server enables any Claude user to govern their agent actions through Sift before execution.

Tools

  • sift_authorize -- get a signed receipt before executing any real-world action
  • sift_check_policy -- check agent permissions
  • sift_register_agent -- onboarding instructions

Why this matters

AI agents that touch money, data, or infrastructure need a governance layer. Sift provides cryptographic pre-execution authorization with full audit trail and fail-closed design.

Website: https://sift.walkosystems.com
Source: https://github.com/walkojas-boop/sift-mcp

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant