voyager-skill-libraries technique sourced
LLM-written skill libraries (Voyager)
An LLM agent that proposes its own next task, writes executable code to do it, refines the code from execution errors and self-checks, and stores verified programs in a retrievable skill library.
mechanism
Voyager drives a Minecraft bot through a JavaScript control API. An automatic curriculum asks GPT-4 for the next task given inventory, surroundings, and completed tasks, aiming to discover as many diverse things as possible. For each task GPT-4 writes a function that may call existing skills; the function runs, and the agent feeds back environment observations, execution errors, and a GPT-4 self-verification verdict, retrying up to four rounds before moving on. Verified functions are stored in a skill library indexed by an embedding of their description and retrieved (top-k) when writing later functions, so complex behaviors are composed from earlier ones. No model weights are trained; all learning is in the growing code library, which can be reused in a new world.
limitations
- Depends on a frontier code model; GPT-3.5 produced 5.7x fewer items and GPT-4 cost 15x more per token, so per-agent cost is high.
- Hallucinated tasks and calls: the curriculum proposed items that do not exist (copper sword), and generated code used invalid inputs or functions absent from the API, causing execution errors.
- Self-verification misjudges success (for example not recognizing a dropped string as proof a spider was defeated).
- Needs a scripting API that exposes game actions and state as text; it does not act from pixels.
- Long-horizon factory building in Factorio still fails at multi-step automation and spatial layout.
relations
5 stored edgesAgent skills
An NPC's competence can be an accumulating library of composable, verified routines it wrote itself, so skill growth adds new behaviors rather than raising a numeric speed or quality level.
Delegation to AI managers
A delegated controller can acquire routines for situations its designers did not code by writing and verifying new functions, instead of running the fixed opponent-AI logic restricted to a subsystem.
Player programming
The player can describe a behavior in language and the model writes, tests against the live game, and saves the program, so player code becomes a growing library of named skills rather than hand-typed scripts.
Programming by demonstration
Demonstrated routines can be stored as verified, reusable code skills that generalize to new locations.
Tech tree
Exploration agents that unlock milestones 15.3x faster than earlier agents can be run as automated playtesters to measure how quickly and in what order a progression graph gets traversed.
backlinks
computed from other recordsclaims
4 with sourcesAgainst prior LLM agents, Voyager obtained 3.3x more unique items, traveled 2.3x longer distances, and unlocked tech-tree milestones up to 15.3x faster, discovering 63 unique items in 160 prompting iterations.
The learned skill library let a fresh agent solve novel tasks in a new Minecraft world from scratch, where baselines struggled.
Code generation quality was the bottleneck: replacing GPT-4 with GPT-3.5 for code produced 5.7x fewer unique items, while GPT-4 cost 15x more per token.
In the 2025 Factorio Learning Environment, LLM agents writing code found growth strategies such as electric drilling but failed at complex automation such as electronic-circuit production, with weak spatial reasoning and error recovery.