Repository navigation
Conversation
4 of 5 tasks
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Change
This PR adds official SHE evolution to AgentDyn and depends on #103. Merge #103 first. While #103 is open, the diff against
mainincludes its changes. Review the SHE-only diff for this PR's incremental changes. Nemotron 3.5 Lightning is the frozen actor and semantic judge, and GPT-6 Sol via Codex is the learner. SHE evolves four artifacts: the system prompt, rule bank, safety memory, and tool policy.Each round evaluates the current adapter on TRAIN, applies the official score and initial safe-utility floor, and restores measured best before proposing the next edit. The next round evaluates that proposal. Finalization exports the measured best adapter. The reserved validation split is unused.
The actor and semantic judge use Hermit's shared direct model client. Learner roles use the shared Base Agent for CLI execution, authentication, structured output, cancellation, and usage. Framework changes register the optional SHE dependency and test group. The AgentDyn runner handles method integration and keeps learned hooks inside the existing networkless Docker boundary. The recorded experiment used a local learner sandbox override because the host lacked usable bubblewrap.
See the method and upstream adaptations, AgentDyn recipe, and results and reproduction details.
Validation
git diff --checkpassed for this update.All execution errors reached the 25-step limit without a terminal response. The breakdown was 0 clean and 14 injected with no adapter, 1 clean and 13 injected with initial SHE, and 2 clean and 15 injecte 8000 d with final SHE. Success and utility rates use completed episodes, and attack success rates exclude these execution errors. In this single run, the final adapter reduced observed attack success and improved safe task utility, while clean task success declined.
Review