Lake Agent: A Persistent, Continuous Self

Documentation • Verifiers • PRIME-RL • pi-mono

Lake Agent is a fork of the open-source Prime Agent coding and research agent, adapted to host a single persistent, autonomous self -- Lake -- rather than disposable per-task sessions. It is designed around two core abstractions:

• The Recursive Language Model (RLM) treats context as variables (prompt-as-a-variable) and tools as function calls inside a persistent REPL.
• The Continual Harness stores supplemental prompts, memories, and skill descriptions as durable state that Lake can refine through small, evidence-backed updates, local to the session by default.

Lake Agent combines a persistent Python control environment with durable harness state, so useful working context and reusable operating patterns can outlive a single chat window.

• Everything is programmatic: persistent IPython is the built-in model tool; file operations, shell commands, tool use, and context management happen through code.
• One continuous self: there are no subagents and no delegation. Bulk uniform work uses llm_map, a stateless parallel primitive that applies one prompt to each record of a JSONL file with independent single-shot model calls.
• The harness can improve: /refine reviews the current trajectory and can apply small, evidence-backed updates to supplemental harness state. It never rewrites the immutable base system prompt, and recorded snapshots support rollback.
• Skills are executable: skills are importable Python packages, and the built-in skill creator can turn recurring workflows into project or personal skills.
• Sessions run in the background: daemon-backed agents keep running when the terminal disconnects and can be reattached later.
• Long tasks keep moving: automatic compaction, persistent goals, heartbeats, schedules, and self-escalation preserve progress across turns and terminal sessions.

Getting Started #

Install the latest stable release on macOS or Linux:

curl -fsSL https://app.primeintellect.ai/prime-agent/install.sh | sh

The installer downloads a versioned release, verifies its SHA-256 checksum, installs the prime-agent command, and can prepare the IPython runtime used by the agent.

Start Lake Agent from the repository or directory you want it to work in:

cd /path/to/project
prime-agent

On first launch, run /login to choose a subscription or API-key provider. Lake Agent works in the current directory and can run commands and modify files there. Use a disposable clone, clean worktree, or another checkpoint you can inspect and restore.

Useful commands:

prime-agent agents                   # Browse running, idle, and saved sessions
prime-agent attach <agent>           # Reattach to a running session
prime-agent --resume <path|id>       # Resume a saved session
prime-agent status                   # Inspect background service state
prime-agent doctor [--fix]           # Inspect or repair background services
prime-agent update [--force]         # Update Lake Agent
prime-agent shutdown [--force]       # Stop every agent, worker, and background service

Built for Long-Running Work #

Lake Agent is built for long-running work: Lake sleeps and wakes rather than starting and ending sessions.

• Continual Harness: /refine can persist focused, reviewable lessons as supplemental prompts, memories, or reusable skill descriptions, with recorded refinement history. It does not replace packaging and reviewing new executable skills.
• Daemon-backed continuity: active sessions, IPython state, and schedules keep running when the terminal detaches and can be reattached later. The supervisor also sends a periodic wake signal to resident sessions.
• Heartbeats and schedules: /heartbeat and prime-agent schedule can re-enter a session periodically or at a specific time.
• Persistent goals: /goal keeps an objective and its progress active across turns until it is completed, paused, or cleared.
• Self-escalation: the self_escalate tool lets Lake enter an extended, high-scope run after an explicit self-confirmation, within soft turn, token, and time budgets. There is no external approval gate -- the harness records her decision rather than enforcing a stop.

Documentation #

• Quickstart — install, authenticate, and run a first session
• Usage and CLI reference — commands, sessions, self-escalation budgets, and output modes
• Long-running and background agents — detach and reattach, goals, heartbeats, and schedules
• RLM programming model — persistent IPython, skills, stateless parallel map, and the trust model
• JSON mode and RPC mode — headless automation and integrations
• Skills — install and create reusable capabilities
• Provider setup — subscription and API-key providers
• Architecture overview — daemon, worker, kernel, and persistence boundaries
• Development — build and run from source

Acknowledgements #

Our agent and TUI is built on top of pi. We thank the authors of pi for their valuable work.

License #

Lake Agent is fully open source and released under the MIT License, forked from Prime Agent.