Repository navigation
feat(observers): break preset trajectory cycle detection and loop governance (#93) - #95
Merged
rosspeili merged 1 commit intoOct 16, 2026
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #93
Description
Enhances the standard
BreakObserver(preset: break) with sliding-window trajectory cycle detection, cognitive deadlock sensors, FinOps token burn guards, and cryptographic non-repudiation proofs while preserving 100% backward compatibility for legacymax_identical_intents.What changed & why:
difflib.SequenceMatcher.max_cycle_token_wasteis exceeded.cycle_entry_hash,cycle_recurrence_hash) inobserver.alertpayloads for verifiable algorithmic deadlock auditing.remedial_directivesteering strings with prompt-level recovery guidance (prefixedBREAK:).action: "alert_only"(default),action: "pause"(appendsbreak_pauserule and trips session state), and full integration withprofile.escalations[](on: observer.alert).scorecard.observers = {alerts, loop_cycles}and findingCOGNITIVE_CYCLE_DETECTED(severity: warn).Type
Checklist
Fixes #93)pytestpasses (300/300 passed)black aura tests— no diff (121 files clean)flake8 aura tests— clean (0 errors, 0 warnings)CHANGELOG.mdupdated when user-visible behavior changesTest plan
pytest tests/test_observer_break_loop.py):python scripts/aura_break_loop_stress_sim.py):pytest):black --check aura tests scripts: 121 files verified clean.flake8 aura tests scripts: 0 errors, 0 warnings.