Operating across tools (Markdown)

Operating mode

# When you're operating across tools

Every session starts from zero. You re-explain context, re-prompt corrections, re-establish what your agent already knew.

You open Claude, Cursor, ChatGPT, or Codex to get work done. Nothing your agent learns in one session is guaranteed to be there in the next. When it gets something wrong, there's no way to correct it that sticks. You're the human sync layer between every tool. Neotoma gives you continuity so you can steer instead of drive.

Escaping

Context janitor — human sync layer between tools

Into

Operator with continuity — steering, not driving

Turn-by-turn prompting → review-and-steer

Tax you pay

Re-prompting, context re-establishment, manual cross-tool sync

What you get back

Attention, continuity, trust in your tools

## Same question, different outcome

Without a state layer, agents return stale or wrong data. With Neotoma, every response reads from versioned, schema-bound state.

Tasks

without state layer

What are my open tasks?

No tasks found.

with state layer

What are my open tasks?

3 open tasks. Next due: submit proposal by Friday.

Task created in Claude, invisible in Cursor

You told Claude to track a deadline. Later you asked Cursor for open tasks. The deadline didn't exist - each tool keeps its own disposable context with no shared memory.

People & contacts

without state layer

Email the latest draft to Priya.

Sent to priya@oldco.com.

with state layer

Email the latest draft to Priya.

Sent to priya@newco.io.

Stale contact, wrong email sent

You updated a contact's email in one conversation. The next session used the old address because the correction was silently lost.

Financial records

without state layer

Find the Whole Foods receipt from Feb 8.

No receipts found matching that query.

with state layer

Find the Whole Foods receipt from Feb 8.

Whole Foods, Feb 8 ($47.32). Stored from conversation on Feb 8.

Receipt stored, then lost

You shared a receipt in a chat. Weeks later you needed it for an expense report. Gone - conversation-scoped memory doesn't keep documents.

Events & commitments

without state layer

Do I have anything due this week?

Nothing scheduled.

with state layer

Do I have anything due this week?

Follow up with Kenji re: proposal, due Thursday.

Commitment forgotten between sessions

You told your AI you'd follow up with a client by Thursday. By Wednesday, neither tool remembered - the commitment was locked in a prior session's expired context.

◆

## Why this happens

×Every session starts from zero; nothing your agent learns carries over

×Context scattered across email, drives, screenshots, and chat histories

×You re-explain the same project, preferences, and constraints in every conversation

×Commitments and action items vanish between sessions and tools

×Your personal data lives in provider memory you don't control

Failure modes without a memory guarantee

Commitments lost between tools

Context breaks when you switch tools

Facts silently drift over time

Corrections don't stick

Personal data in provider memory with no deletion control

Memory locked to one vendor

### Every session starts from scratch

You re-explain the same project context, preferences, and constraints in every new conversation. Memory doesn't follow you across tools, and it silently drifts as models compress or discard what you told them.

### Commitments vanish between tools

You tell Claude to remind you about a deadline. Later you ask Cursor for open tasks. The deadline doesn't exist. Action items created in one session have no guarantee of surviving to the next.

### Corrections don't persist

When your AI gets a date wrong, associates the wrong contact, or misidentifies something, you can't tell it "that's wrong" in a way that lasts. The mistake reappears next session.

### Your data in someone else's memory

Your receipts, contacts, health information, and financial records live in provider-hosted memory. No transparency into retention, no guarantee against training use, no delete button.

◆

## AI needs

What you need from your AI tools, and what current tools don't provide.

- [Memory that survives](/memory-models#deterministic-memory) session resets and tool switches
- [One source of truth](/foundations#cross-platform) across Claude, Cursor, Codex, and ChatGPT
- Automatic extraction of commitments, tasks, and contacts from conversations
- [Corrections that stick](/deterministic-state-evolution): fix once, fixed everywhere
- [Every stored fact traces back](/auditable-change-log) to where it came from

◆

## How Neotoma solves this

Neotoma removes the tax you pay re-explaining your world to every tool. Store a fact once and it's available everywhere — Claude, Cursor, Codex, ChatGPT. Correct once and it sticks.

[

Shared memory across every tool

Every agent connected to Neotoma reads from and writes to the same memory. Store a task in Claude, retrieve it from Cursor. One memory, not one per tool.

](/foundations#cross-platform)

Facts extracted automatically every turn

People, tasks, events, preferences, and commitments are extracted from every conversation turn and stored before the agent responds.

[

Corrections that stick

Correct something once. The fix persists across every tool and session. Same question, same answer, every time.

](/deterministic-state-evolution)[

Full history with provenance

Every conversation and fact is stored with its source. See what was known at any point in time and where each fact came from.

](/versioned-history)

◆

## What actually changes

Without persistent memory, you drive every turn. Every prompt carries the full weight of what came before because the system won't hold it for you.

With Neotoma, the agent arrives at each session already knowing what it knew last time. Your role shifts from re-explaining your world to reviewing what the agent knows and correcting when it's off.

Fewer prompts, shorter sessions, more done. Not "let me re-explain my situation" - "here's what changed since yesterday."

◆

## Data types for better remembrance

The entity types you'll store most often.

conversation

Chat sessions with full turn history across tools

message

Individual turns with role, content, and extracted facts

task

Commitments, reminders, and action items with status and deadlines

note

Captured thoughts, observations, and reference material

contact

People and their details (email, role, organization)

event

Calendar events, deadlines, and commitments

preference

Settings and preferences that persist across sessions

receipt

Purchase records, invoices, and expense tracking

◆

## When you don't need this

For one-off questions or single-document analysis, your AI tools already work fine. Neotoma is for when you need what you told one tool to still be true when you open another - and for knowing that a correction you made actually stuck.

◆

## Other modes

The same person operates in multiple modes. The tax differs; the architecture that removes it is the same.

[

Building mode

Building pipelines

Your agent guesses entities every session. Corrections don't persist. Memory regressions ship because the architecture can't prevent them.

](/building-pipelines)[

Debugging mode

Debugging infrastructure

Two runs. Same inputs. Different state. No replay, no diff, no explanation.

](/debugging-infrastructure)

◆

The tax is re-prompting, re-explaining, and manually syncing context between tools. Neotoma removes that tax and gives you back the attention and continuity it was consuming.

Built by someone who runs every workflow (email, finance, content, tasks) through the same multi-tool stack.

Deep dive: [Agentic retrieval infers. It doesn't guarantee.](https://markmhendrickson.com/posts/agentic-search-and-the-truth-layer)

[Install in 5 minutes](/install)[View architecture →](/architecture)