personal-model 0.3.2

Local-first macOS Runtime for screen-context memory, personal modeling, and MCP.

Persome

The local-first Personal Model Runtime for macOS. Persome observes the apps you already use, turns cross-app activity into an inspectable model of a real person, and serves that model to MCP agents.

Star Persome on GitHub to follow the Runtime and help prioritize the next MCP integrations.

Actual /model screenshot produced by scripts/sample_demo.py --showcase: 424 synthetic Points, 146 Lines, 12 Faces, 4 Volumes, and 1 Root. It contains no personal data.

Product job

Persome runs quietly on one Mac and does four jobs:

1. Collect focused macOS Accessibility (AX) context across apps, with an optional on-device OCR fallback for AX-poor surfaces.
2. Model observations into sourced facts, evolving relations, stable patterns, cross-domain structure, and one current Root.
3. Serve local memory and model tools over MCP.
4. Give control back through receipts, time travel, correction, export, and deletion.

This is the Runtime, not a hosted account or a single assistant's private memory. One local model can be used by Claude Code, Codex, Cursor, or another trusted MCP client.

Five-minute sample demo

See the whole model without an API key, Accessibility permission, or access to your real ~/.persome data. This path requires Git and uv:

git clone https://github.com/Intuition-Lab/personal-model.git
cd personal-model
uv run python scripts/sample_demo.py

Add --showcase to render the denser, still fully synthetic model used in the README image.

The script opens http://127.0.0.1:8743/model, serves MCP at http://127.0.0.1:8743/mcp, and deletes its temporary synthetic data when you press Ctrl-C. To inspect the exact search, receipt, and snapshot payloads:

PERSOME_LLM_MOCK=1 uv run python scripts/sample_demo.py --json

With the sample server still running, verify the actual MCP transport from a second terminal:

uv run python scripts/verify_sample_mcp.py

This sample path is deliberately separate from the real-data path below.

Quick start with your data

Requirements: macOS 13 or newer, Xcode Command Line Tools, and a Python build with SQLite 3.42+ (the installer verifies the secure FTS capability). The installer finds or installs uv, provisions Python 3.11-3.13, compiles the Swift AX helpers into immutable source-versioned paths under ~/.persome/native/, generates the local screenshot-encryption key, configures local OCR, and offers to register detected MCP clients. Before it reports success, onboarding explains and requests Accessibility for the capture helper and event watcher separately, then requests Screen Recording only when the effective pixel policy needs it. It starts Persome, proves the final lifecycle owner and Runtime generation, and obtains a mode-appropriate capture receipt. Its fallback uv download is version-pinned and checked against repository-pinned SHA-256 digests; the Runtime environment is installed from the committed uv.lock, and the complete build-backend closure is hash-constrained rather than resolved afresh.

Install the published PyPI distribution with uv and run the same explicit onboarding proof:

uv tool install personal-model
persome onboard

The distribution is named personal-model; the installed CLI remains persome. The source installer below remains the most explicit first-run path and is also used by transactional updates.

Upgrade a PyPI/uv tool installation with its package manager, then re-run the same Runtime proof:

uv tool upgrade personal-model
persome onboard

git clone https://github.com/Intuition-Lab/personal-model.git
cd personal-model
bash install.sh

persome status
persome model open

# Repeat only to repair/recheck permissions and Runtime proof:
persome onboard
# Diagnostic; exits nonzero for intentionally unconfigured optional features:
persome doctor

persome onboard is the repeatable recovery path. In standard daemon-capture mode it presents a plain-language explanation before each macOS request. The versioned mac-ax-helper and, when event-driven capture is enabled, mac-ax-watcher are the Accessibility principals; the terminal and Python daemon are not substitutes for those grants. On Apple Silicon it also verifies the isolated OCR worker when OCR is enabled, leaves the daemon running, and forces one fresh capture through the daemon-owned runner. The authenticated /permissions result re-runs the actual helper/watcher probes and the Runtime's Screen Recording preflight. If HTTP auto-start is intentionally disabled, the same generation publishes equivalent owner-only readiness and capture receipts in .runtime-state.json. Trusted ingest, Intel without local OCR, explicit OCR/pixel opt-out, and paused/locked update flows report their effective mode rather than claiming an OCR capture occurred. OCR supplies text for AX-poor apps such as WeChat and Feishu; pixels never enter an LLM prompt. The first OCR model load can take up to two minutes; repeated onboarding normally finishes in seconds without warming a second worker. Completion never waits on a hidden final dialog. Persome does not require Full Disk Access.

OCR intent is durable in [capture].ocr_policy: auto is an unconfigured fresh install, while enabled and disabled record an explicit choice. Ordinary persome onboard preserves an existing tier or opt-out. Running persome onboard --tier tiny or persome ocr setup explicitly enables OCR; persome ocr disable records the opt-out. In source="ingest" mode the trusted producer owns macOS capture permissions and Persome proves the authenticated ingest runner instead of pretending that its daemon produced an AX frame.

Update an existing installation

For an installation created by install.sh, run the transactional updater from any directory; no Git checkout is required:

persome update

The command downloads a fresh shallow checkout of the official main branch and builds a relocatable venv.replacement.update while the active venv remains untouched. Activation rejects a console script that still embeds the inactive candidate path. Under one owner-only update lock it stops the old Runtime, atomically exchanges the prepared and active directories, restores the prior lifecycle owner, and runs mode-aware onboarding against that exact generation. Only a successful permission, policy, owner, health, and capture/readiness proof commits the exchange. A failed, interrupted, or crash-recovered update atomically exchanges the old virtualenv back and restores its Runtime owner; repeated interrupts cannot abort recovery. Configuration, credentials, personal data, capture policy, and lifecycle intent under ~/.persome are preserved, and a developer checkout is never modified. To install an already-reviewed local tree instead:

persome update --source /path/to/personal-model

Running bash install.sh again over an existing installation automatically delegates to this same transactional update path; it does not replace a live virtualenv in place or rerun LLM provider setup.

The AX binaries are keyed by their source bytes and architecture. Reinstalling the same version reuses the exact executable (and therefore its existing macOS grant); a release that changes helper source uses a new path and onboarding asks for that new principal explicitly. If an update rolls back, the old Runtime resolves the old helper path again.

# Recheck or repair OCR onboarding; disable is always explicit and reversible.
persome onboard
persome ocr setup
persome ocr status --check
persome ocr disable

An LLM is optional for collection and BM25 recall, but required for semantic modeling. During installation, the provider wizard asks you to choose a service and enter its API key. Persome supplies that provider's endpoint and default model, tests completion and tool calling, and only then saves the route. Existing keys are detected automatically. API keys go to the owner-only ~/.persome/env file under the provider-neutral PERSOME_LLM_API_KEY name; provider-specific environment variables are import sources only. The non-secret route goes to ~/.persome/config.toml. Nothing ships with a key.

# If provider setup was skipped during installation:
persome llm providers
persome llm setup
persome llm status --check

# Restart after changing the active provider:
persome stop || true
persome start

Persome speaks two wire protocols: native Anthropic Messages and OpenAI-compatible Chat Completions. Presets cover Anthropic, OpenAI, DeepSeek, OpenRouter, Gemini, Groq, Mistral, xAI, Qwen, Moonshot/Kimi, Zhipu GLM, SiliconFlow, Together, Fireworks, Cerebras, Azure OpenAI, Ollama, LM Studio, and vLLM. custom-openai and custom-anthropic accept another compatible endpoint. Azure and custom endpoints use a clearly marked advanced setup path. A preset means the route is configured, not that every model has the necessary capabilities; Persome warns when the default model cannot call tools.

Active work is reduced every five minutes by default. A first useful recall is therefore expected within ten minutes of valid capture plus a working semantic provider; persome status, persome model status, and the viewer explain sparse or degraded states instead of inventing geometry.

Proof points

Local-first

• Durable Markdown, SQLite/FTS5, model snapshots, and logs live under ~/.persome unless PERSOME_ROOT is set.
• AX is the default signal. Optional PP-OCRv6 runs locally in an isolated subprocess with bundled weights.
• The HTTP/MCP server is restricted to loopback (127.0.0.1 by default), requires an owner-local bearer on API/MCP routes (or its one-use derived viewer capability), and emits no telemetry.
• Only configured semantic stages send derived text to the selected provider's LLM or embedding endpoint.

Cross-app

The source-versioned Swift watcher notices AX events and the matching helper reads the focused AX tree across native and browser apps. Persome normalizes focused element, visible text, window, application, URL, and time into one capture and session pipeline. OCR is a fallback, not a parallel cloud recorder.

Agent-ready

• Authenticated streamable HTTP MCP: http://127.0.0.1:8742/mcp
• stdio MCP: persome mcp
• Stable model contract: persome model export and GET /model/graph
• Evidence tools: search, read_receipt, verify_fact, and get_model_snapshot

Connect an MCP client

Register an owner-local stdio server. These clients launch it on demand, so the daemon does not need to be running and no bearer is copied into their config:

persome install claude-code
persome install codex
persome install claude-desktop
persome install opencode

# Generate a stdio config that can be merged into Cursor's MCP config:
persome install mcp-json --filename persome-mcp.json

Client	Verified configuration	Check
Claude Code	persome install claude-code	claude mcp list
Codex CLI / IDE	persome install codex	codex mcp list
Claude Desktop	persome install claude-desktop	fully quit and reopen the app
opencode	persome install opencode	opencode mcp list
Cursor	merge the generated mcpServers.persome object into .cursor/mcp.json or ~/.cursor/mcp.json	Cursor Settings -> MCP

The canonical JSON shape is:

{
  "mcpServers": {
    "persome": {
      "command": "persome",
      "args": ["mcp"]
    }
  }
}

See MCP client setup and verification for authenticated HTTP configs, uninstall commands, and privacy boundaries.

Real MCP query with a cited answer

The following result is generated by the committed synthetic sample through the same search and read_receipt implementation exposed by MCP.

Tool: search
Input: {"query":"When does the user prefer focused writing?","top_k":2}

Top result:
  id:        20260701-0800-d4e5f6
  path:      project-work.md
  timestamp: 2026-07-01T08:00
  content:   The user reserves mornings for focused writing and review.

Tool: read_receipt
Input: {"entry_id":"20260701-0800-d4e5f6"}

A grounded client response can then say:

The user prefers mornings for focused writing and review. [project-work.md, 2026-07-01 08:00; receipt 20260701-0800-d4e5f6]

The receipt is resolvable, the superseded earlier statement remains available as history, and the answer does not rely on the model's unsupported memory.

Benchmark and verification status

This repository reports Runtime engineering evidence, not a paper-quality personalization benchmark.

Gate	Public evidence	Current status
Fresh root -> complete geometry	tests/test_runtime_model_e2e.py	deterministic synthetic pass
MCP search -> receipt	sample_demo.py + verify_sample_mcp.py	real streamable HTTP MCP, deterministic synthetic pass
Offline Runtime behavior	pytest -m "not macos and not integration"	complete offline suite; no provider key
Package completeness	clean wheel install + bundled Swift, Three.js, and PP-OCRv6 checks	required by CI/release
Release provenance	SHA-256 manifest + GitHub artifact attestations from a tag reachable from main	required by release workflow
Secret and personal-data safety	secret_scan.py + pii_scan.py	required by CI/release
Memory quality / next-action prediction	separate benchmark repository	not reported here

The sample uses synthetic fixtures and cannot establish recall quality on a real person. No cross-user benchmark, next-action accuracy, latency percentile, or comparison win is claimed. The launch machine's three isolated source installs had an 11.896-second median with a warm uv cache; conditions and limitations are recorded in benchmark scope.

Why Persome

These projects solve adjacent but different jobs:

System	Primary job	Where Persome differs
screenpipe	searchable local screen/audio history and developer platform	Persome centers an evolving Point/Line/Face/Volume/Root personal model with correction and receipts for MCP agents.
Mem0	a memory layer populated by application or conversation events	Persome begins with ambient macOS work context, owns the local capture/session pipeline, and exposes an inspectable model rather than only a memory API.
Assistant/platform memory	convenience inside one provider or client	Persome is a local Runtime shared across trusted MCP clients; data, export, correction, and deletion remain under the user's control.

Persome is not a replacement for a full screen archive, a hosted vector memory, or a provider's preference feature. Choose it when the core requirement is a local, cross-app, auditable model that multiple agents can query.

How it works

flowchart LR
  AX[macOS AX watcher] --> S0[S0 debounce]
  OCR[Optional local OCR] --> S1[S1 normalized capture]
  S0 --> S1
  S1 --> BUF[Capture buffer]
  BUF --> TL[1-minute timeline]
  TL --> SES[Deterministic sessions]
  SES --> DELTA[5-minute memory delta]
  DELTA --> PL[Points and Lines]
  PL --> FV[Faces and Volumes]
  FV --> ROOT[Root]
  PL --> RET[BM25 and optional dense retrieval]
  FV --> MCP[MCP, export, viewer]
  ROOT --> MCP
  RET --> MCP

Every modeled object keeps source receipts and bitemporal history. A sparse store can truthfully contain Points and Lines without a Face, Volume, or Root. The viewer shows that incomplete state rather than fabricating one.

Read Runtime architecture, the model contract, and the detailed maintainer architecture.

Inspect, correct, export, and delete

# Inspect
persome status
persome model status
persome faces-report
persome contradictions
persome model open

# Correct or revoke one memory while retaining its audit trail
persome correct --help
# Agents can also call MCP correct_memory.

# Export a redacted owner-only snapshot (0600)
persome model export

# Delete model memory, or all captures/timeline/model state
persome stop
persome clean memory
persome clean all

For a complete uninstall that preserves personal data by default:

bash uninstall.sh

# Explicitly remove the remaining data, config, env, exports, and logs:
bash uninstall.sh --delete-data --yes

Client registrations are removed separately and idempotently:

persome uninstall claude-code
persome uninstall codex
persome uninstall claude-desktop
persome uninstall opencode

See operations and data control for exact paths, backup advice, export sensitivity, reset behavior, and manual removal steps.

Privacy boundary

• Personal data remains local until a configured model stage or connected agent sends selected text to its own provider.
• MCP capture tools can return raw screen text, titles, URLs, and focused-field values. Bearer/stdio access is a personal-data capability; connect only clients you trust.
• Screenshots are omitted from MCP by default and encrypted at rest when retention is enabled.
• persome model export is redacted by default; --raw is an explicit opt-out.
• There is no built-in remote account, sync service, telemetry, meeting audio capture, computer-use actuation, or filesystem profiler.

Read Security and privacy before using real personal data, and report vulnerabilities through SECURITY.md.

Platform support

Platform	Capture	Local OCR	Runtime / MCP
macOS 13+ on Apple Silicon (arm64)	supported	bundled PP-OCRv6	supported
macOS 13+ on Intel (x86_64)	supported AX path	unavailable because Paddle does not ship the required Intel wheel	supported
Linux	no live macOS capture	not packaged	offline tests and development only
Windows	unsupported	unsupported	unsupported

Python 3.11-3.13 with SQLite 3.42+ is supported by the installer. See operations and troubleshooting.

Persome and Personome

Persome is this open-source Runtime and project name. Personome is the research term for the learned model of one person: a dynamic state assembled from sourced observations, relations, stable patterns, and higher-level structure. The product name stays Persome in commands, packages, paths, APIs, and documentation.

Paper and architecture-note status

This repository ships the executable Runtime and an implementation-oriented architecture note. The architecture documents are not a peer-reviewed paper, and the Runtime's synthetic gates are not publication benchmarks. The paper, benchmark suite, data statements, and project publication will live as separate artifacts with independent licenses before release. See licensing boundaries and benchmark limitations.

Roadmap

The public roadmap is issue-driven:

• more tested MCP client integrations;
• richer first-run permission diagnostics;
• explicit import/export interoperability;
• Intel and future-macOS compatibility evidence;
• a separate, reproducible personal-model benchmark suite.

Browse starter issues or start a design question in Discussions.

Contributing and community

Read CONTRIBUTING.md, follow the Code of Conduct, and use SUPPORT.md to choose the right channel. Every commit requires DCO sign-off, and CI blocks known secrets, personal data, non-English source text, contract drift, lint failures, and offline regressions. Third-party Actions are pinned to reviewed commit SHAs and workflow permissions default to read-only.

Contributors

Persome is shaped by people across engineering, design, research, and community.

Singularity   💻 Code	Li_Xufeng   💻 Code	Siyi   🎨 Design
Kevin   💻 Code	huachenjie238-oss   📈 Growth	Jing@Meowy   📈 Growth
Zhiheng Chen   💻 Code

_{Contribution labels follow the All Contributors convention. Contributions of every kind are welcome.}

Support Persome

If an inspectable, user-owned personal model is useful to your agents, star Persome on GitHub and share the MCP client or workflow you want supported in Discussions.

License

Runtime code is Apache-2.0. Paper, benchmark, project-note, third-party, and personal-data boundaries are explained in LICENSES.md. Required incorporated-work notices remain in NOTICE and THIRD_PARTY_NOTICES.