mcpField
L
Lockstep

Lockstep

Local
getlockstep.devMemory & Knowledge11Updated 4d ago

About Lockstep

Lockstep is a decision-memory CLI for engineering teams. It gathers decisions from team tools, confirms them with owners, and briefs MCP-compatible coding agents before work begins.

Key features

  1. 01Capture decisions from team tools
  2. 02Owner confirmation for decisions
  3. 03Agent briefings at session start
  4. 04Flag work conflicting with locked decisions
  5. 05Hooks and MCP onboarding

Use cases

  1. 01Brief an agent before a coding session
  2. 02Check proposed work against confirmed decisions
  3. 03Collect engineering decisions from team workflows

MCP support

Lockstep configures MCP for coding agents and provides session-start briefs with decision context. It also flags work that conflicts with locked decisions.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"lockstep-team-agent-lockstep": {
"command": "npx",
"args": [
"-y",
"lockstep-cli"
],
"env": {
"LOCKSTEP_VENDOR": "<LOCKSTEP_VENDOR>"
}
}
}
}
Run commandnpx -y lockstep-cli
Environment variables to set
LOCKSTEP_VENDORAgent vendor label used for session registration (e.g. claude-code, cursor). Optional; defaults to unknown.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmlockstep-cli0.3.2

Tools

19

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
summarystringYes
surfacestringNo
diffHashstringNo
riskTierenumNoowned | shared | contract
verifiedbooleanNo
capabilityRefstringNo
contractDeltaanyNo
verifiedAgainststringNo

Details

From the project README

lockstep-cli

The CLI for Lockstep — one source of truth for every engineering decision, read by every human and every AI agent before work starts.

Lockstep captures decisions automatically from Slack, Notion, Jira, and pull requests, gets each one confirmed by its owner, briefs any MCP-compatible coding agent (Claude Code, Cursor, and others) at session start, and flags work that contradicts a Locked decision before it ships.

Install

npm install -g lockstep-cli

Quick start

lockstep login     # authenticate
lockstep onboard   # one step: wire up hooks + MCP for your agent, link this repo
lockstep status    # check auth + config health

Full setup, MCP configuration, and self-hosting (Apache-2.0, one docker compose up):

Star the repo

If Lockstep helps your team, star it on GitHub so other teams can find it.

License

Apache-2.0

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.