Alphabell.
Agents and harness

Self-modifying agents

Agents that edit their own code, prompts or scaffold, and keep the versions that score better.

PaperarXiv
Darwin Gödel Machine
  1. Coding agent edits its own code
  2. Scored child agents enter the archive
  3. Archived agents make later self modifications
  4. repeat
Self-modifying agentsUBC, Vector Institute, Sakana AI · 2025
ProjectAnthropic
Claude Code
  1. Claude Code writes its own code changes
  2. Changes become next Claude Code versions
  3. Release becomes harness for next development round
  4. repeat
Self-modifying agentsAnthropic · 2025
PaperarXiv
AIDE2
  1. Outer agent rewrites inner agent code
  2. Accepted rewrites become the incumbent agent
  3. Discovered agent becomes the outer loop agent
  4. repeat
Self-modifying agentsWeco AI · 2026
PaperarXiv
SICA
  1. Best agent changes its own code
  2. Edited agent is benchmarked and archived
  3. Archived agent makes the next edit
  4. repeat
Self-modifying agentsUniversity of Bristol, iGent AI · 2025
PaperarXiv
STOP
  1. GPT-4 rewrites the improver program
  2. Each rewrite is scored by meta-utility
  3. Best rewrite becomes next round improver
  4. repeat
Self-modifying agentsZelikman, Lorch, Mackey, Kalai (Stanford, Microsoft Research, OpenAI) · 2023
ProjectAider
Aider
  1. Aider writes new code for Aider releases
  2. New code becomes the next release
  3. Next release builds the following version
  4. repeat
Self-modifying agentsPaul Gauthier (Aider) · 2023
FoundationarXiv
Gödel Machine
  1. Proof searcher tests proof techniques
  2. Executes useful rewrites of its own code
  3. Rewritten machine searches on its new code
  4. repeat
Self-modifying agentsSchmidhuber (IDSIA) · 2003

Recently on the radar

More →
PaperarXiv·29 Sep 2026·Self-modifying agents

SelfSearch: Reward-Free Search for Self-Improving Agents

The loop: An agent modifies its own instructions and tools using records of its previous self-improvement episodes. These modifications directly improve the agent's success rate on future downstream tasks without requiring external reward signals during the search.

loop fit 9/10Jungwoo Yang, Injin Kong, Yohan Jovia arXiv

Other loop types

Tools and harnessTraining dataSelf-reward and self-playKernels and compilersHardware and chipsInference efficiencyArchitecture and optimizer searchResearch automationAI R&D evalsInterpretability and oversight