The Lockstep blog
Keeping humans and AI coding agents in step: decision capture, context engineering, MCP, drift, and the record that files itself.
- 7 min read
Best memory tools for Claude Code: choose the missing job, not another memory layer
Compare native Claude Code memory and six tools by the job you need, sharing boundaries and documented limits.
Read → - 6 min read
The CTO's playbook for AI coding agents at a 30-70 person company
At 30-70 people, informal coordination stops working right as agents amplify whatever your team has written down. The decisions to make once, the short list of things to record, and a rollout order that survives contact with real teams.
Read → - 5 min read
RFC template for engineering teams (+ design doc)
A copy-paste engineering RFC and design doc template in Markdown, with Confluence notes, a review process, and how an RFC becomes a recorded decision.
Read → - 5 min read
Notion alternatives for tracking product decisions (2026)
Notion decision logs go stale because every entry is hand-filed. Six alternatives for tracking product decisions, picked by the reason you're leaving Notion.
Read → - 6 min read
How to link decisions to Jira tickets: 5 options compared
Five ways to track decisions in Jira and keep them attached to tickets and PRs: issue links, Confluence pages, Linear docs, ADRs, and automatic capture.
Read → - 9 min read
Decision log template (free) + 5 filled-in examples
A free, tool-agnostic decision log template in Markdown, spreadsheet and Notion form, plus five filled-in examples, including a superseded decision.
Read → - 7 min read
DACI decision-making template (+ DACI vs RAPID)
A free DACI template for Confluence or Notion, a worked example, a DACI vs RAPID vs RACI comparison, and the mistakes that make DACI fail.
Read → - 6 min read
Confluence decision log template (free) + ADR template
A free Confluence decision log template: one page per decision, a self-updating register built with Page Properties, plus an ADR template. Copy, paste, done.
Read → - 7 min read
ADR template: Nygard, MADR and Y-statement (+ examples)
Copy-paste architecture decision record templates (Nygard, MADR 4.0, Y-statement), a repo layout with adr-tools commands, and three worked ADR examples.
Read → - 7 min read
Why coding agent memory files go stale - and what teams should do instead
Memory files turn dangerous when old tasks, proposals, and decisions still look current. Scope, owners, status, and replacement links stop the drift.
Read → - 7 min read
Context engineering for coding agents: what belongs in the prompt, files, and team memory
Coding agents need different context at different times. A practical map for prompts, repository files, task handoffs, live tools, and shared team decisions.
Read → - 13 min read
AI coding agent memory: why agents forget, and the fix
Why AI coding agents forget between sessions, the five memory layers that fix it (rules files, auto memory, handoffs, MCP, decisions), and how to choose.
Read → - 9 min read
Claude Code keeps forgetting context? What to save where
Resume, CLAUDE.md, auto memory, handoffs, plugins, or MCP? How each Claude Code memory option works, what it can't do, and where team decisions belong.
Read → - 8 min read
ADRs for AI coding agents: make every agent read them
Architecture decision records only help AI coding agents when the right record enters context at the right time. A setup for Cursor and Claude Code.
Read → - 8 min read
Best MCP Memory Servers for AI Coding Agents (2026)
Reference Memory, Basic Memory, Mem0 and Graphiti compared as MCP memory servers for Claude Code and Cursor teams, and when you need decisions, not memory.
Read → - 6 min read
CLAUDE.md vs agent memory: what each should store
CLAUDE.md, auto memory, task handoffs, and shared team decisions solve different context problems. Use this guide to decide what belongs where.
Read → - 7 min read
How to share memory between Cursor and Claude Code
Cursor and Claude Code don't share memory. How to move rules, MCP servers, task state, and decisions between them, or switch tools without losing context.
Read → - 5 min read
Cursor keeps losing context mid-project. How to stop re-explaining your codebase
If Cursor loses context after a long session or project switch, separate working context, stable rules, and team decisions so the next agent can continue.
Read → - 7 min read
Why Claude Code forgets architecture decisions - and how teams stop re-explaining them
Claude Code can understand a codebase and still lose the reasoning behind architecture decisions. What to save, where, and how to keep agents aligned.
Read → - 7 min read
Who wrote this code? AI provenance starts with the wrong question
A commit shows who changed a file and an attestation shows how it was built. Neither explains why an AI agent made the change or which decision it honored.
Read → - 10 min read
Two AI agents, one codebase: keeping parallel work in sync
Parallel coding agents can each make a reasonable change and still leave the product inconsistent. Worktrees, handoffs, and shared decision memory fix it.
Read → - 10 min read
How we distill decisions from Slack noise
Inside Lockstep's distillation funnel: a four-test rubric, tiered models, prompt-cached extraction, and a human gate that nothing skips.
Read → - 9 min read
How cross-vendor agent capture works: hooks, MCP, and instruction files
Every major coding agent exposes the same three surfaces. How Lockstep uses lifecycle hooks, an MCP server, and instruction files to capture agent work.
Read → - 9 min read
8 Ways to Give AI Coding Agents Your Team's Context (2026)
AGENTS.md vs CLAUDE.md vs Cursor rules vs MCP vs RAG vs decision memory: what each carries into Claude Code or Cursor, where it breaks, what to pick.
Read → - 9 min read
9 Best MCP Servers for Engineering Teams, Ranked (2026)
GitHub, Playwright, Sentry, Context7, Linear, Slack and 3 more MCP servers ranked for dev teams: what each exposes, auth model, and which to install first.
Read → - 9 min read
Best decision log software for product & engineering teams (2026)
Seven decision log tools compared: Notion, Confluence, Lockstep, ADR files, Linear, GitHub Discussions, and Loomio. Capture, structure, price, and fit.
Read → - 12 min read
The load-bearing hallway: what AI coding agents break first
Git, Jira, and PRDs were never the coordination layer — the human hallway was. AI coding agents delete it. What hospitals and remote teams teach us.
Read → - 9 min read
The engineering decision log: what to write down, and when
A practical guide to keeping a decision log: one question that tells you what counts, the four fields every entry needs, and the two moments teams always miss.
Read → - 16 min read
Architecture Decision Records (ADRs): a practical guide
What an ADR is, the minimal template, and when to write one. Then why ADRs fail on the read side, and what changes when AI coding agents are the reader.
Read → - 16 min read
AGENTS.md: What It Is, Example File, and What to Leave Out
AGENTS.md is a markdown file of instructions for AI coding agents. What it is, which tools read it, a complete example file, and what to leave out.
Read → - 7 min read
MCP vs RAG: When to Use Each
MCP and RAG both get outside information to a model, but they solve different problems. A plain comparison, a decision table, and when to use each.
Read → - 7 min read
The Feed Is Coming for GitHub
The friend graph, then the CRM, now GitHub: value keeps moving from the database up to the layer that reads across everything and decides what happens next.
Read → - 5 min read
Introducing Lockstep: A System of Record for AI Coding Agents
The tools that coordinate engineering work assume humans writing code. AI coding agents have no hallway conversations. Lockstep is the layer they're missing.
Read →