feat: add failure taxonomy and LLM judge template#1
Merged
Conversation
Based on Hamel Husain's evals-skills approach and team input: ## Failure Taxonomy (v1.1) 8 primary modes: MISREAD_SPEC, HALLUCINATION, IMPL_BUG, REGRESSION, TIMEOUT_STUCK, BUDGET_EXCEEDED, CONSTRAINT_VIOLATION, INFRA_FAILURE 7 secondary signals: scope_drift, premature_victory, instruction_amnesia, tool_fumbling, loop_stuck, state_drift, retry_violation ## LLM-as-Judge Template (challenges/_templates/judge.yaml) - Binary 0/1 criteria with weights - Anchored examples (good/bad) - JSON-only output with schema validation - temperature=0, seed=42 for reproducibility - Calibration set for drift detection Inspired by: https://hamel.dev/blog/posts/evals-skills/ Co-authored-by: Loyd <loyd@swarm.local> Co-authored-by: Liselott <liselott@swarm.local>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Based on Hamel Husain's evals-skills approach and team discussion.
Failure Taxonomy (v1.1)
8 primary modes:
MISREAD_SPEC- Task misunderstandingHALLUCINATION- Referenced non-existent thingsIMPL_BUG- Implementation bugREGRESSION- Fixed problem, broke something elseTIMEOUT_STUCK- No progress, loopedBUDGET_EXCEEDED- Token/time/cost limitsCONSTRAINT_VIOLATION- Broke rulesINFRA_FAILURE- Environment failed7 secondary signals: scope_drift, premature_victory, instruction_amnesia, tool_fumbling, loop_stuck, state_drift, retry_violation
LLM-as-Judge Template
challenges/_templates/judge.yamlwith:temperature=0,seed=42for reproducibilityNext Steps
opengym judge <attempt>commandCo-authored-by: Loyd, Liselott (swarm team)