Every commit scored for change-risk against this repo's own history, so 'elevated' means elevated here rather than on some global curve.
Needs review
56 commits sit in this repo's top risk tercile, which is 33% of the 168scored. The cut is drawn against this codebase's own history rather than a global curve, so a quiet repo still fills its top band, and here it starts at 9.0 out of 10. What pushes a commit up is size and spread together: a large change confined to one area scores below a smaller one scattered across a dozen files.
Commit categories over time, read off the subject line. Fixes carry the accent because that is the series this chart exists to show.
Consistently fix-driven across its history.
Ranked by change-risk, highest first. Priority is a tercile of this repo's own distribution, so a quiet repo still fills its top band.
| # | Commit | Author | When | Lines | Risk | Top driver |
|---|---|---|---|---|---|---|
| 1 | 0b74b84fPhase 8: AI Operating System - meta rewards, causal reasoning, swarm intelligence, agent builder, strategy switching, lifelong learning (277 tests, 71 exports, 22 CLI commands) | Kaushalt2004 | 3mo ago | +22.4K -175 | 100%Elevated | more lines added than baseline |
| 2 | 9a507e83Implement advanced memory features: VectorDB, Extractor, Cross-agent, Sleep Cycle | Kaushalt2004 | 3w ago | +37.0K -541 | 99%Elevated | more lines added than baseline |
| 3 | f0f048abfeat: Agent Immune System + Replay Time Travel with RL | Kaushalt2004 | 2mo ago | +4.5K -314 | 99%Elevated | more lines added than baseline |
| 4 | e80e4f41Update PyPI name, versions, and add professional LongMemEval benchmarks | Kaushalt2004 | 1mo ago | +3.3K -7 | 97%Elevated | more lines added than baseline |
| 5 | b1fb3289v0.9.2: update version, PyPI description, add launch assets | Kaushalt2004 | 2mo ago | +3.6K -21 | 97%Elevated | more lines added than baseline |
| 6 | d7ec2495feat: Enterprise Integration Layer - GitHub Issues, CI Fixer, Slack, Linear, Scheduler, PR Reviewer + Task Queue + Setup Wizard + CLI commands | Kaushalt2004 | 2mo ago | +2.5K -4 | 97%Elevated | more lines added than baseline |
| 7 | 9e51badbfeat: Cognitive AI Testing Lab - real code debugging, safety classification, math reasoning demo with live WebSocket streaming | Kaushalt2004 | 2mo ago | +1.8K -0 | 97%Elevated | more lines added than baseline |
| 8 | 8d1da176feat: NEXUS Live Runtime v0.8.0 — Full observability dashboard with 9 real subsystems | Kaushalt2004 | 2mo ago | +2.2K -114 | 94%Elevated | more lines added than baseline |
| 9 | faf26d13feat: real ML agents - DeepQN, PolicyGradient, RandomForest, SVM, XGBoost (ZERO LLMs) | Kaushalt2004 | 2mo ago | +1.5K -1 | 94%Elevated | more lines added than baseline |
| 10 | 9c50e98bAdd 10 missing test files and docstrings for public functions | copilot-swe-agent[bot]copilotautonomous | 3mo ago | +1.2K -0 | 94%Elevated | more lines added than baseline |
| 11 | d1c7bbbbfeat: 4 real GitHub bots - CodeGuardian, IdeaBot, SecurityBot, HealthBot, TestBot | Kaushalt2004 | 3mo ago | +1.0K -0 | 94%Elevated | more lines added than baseline |
| 12 | 482b9aebFix all ruff E,F,W violations: format code, fix unused vars/imports, config E501 handling | copilot-swe-agent[bot]copilotautonomous | 3mo ago | +1.9K -848 | 94%Elevated | more lines added than baseline |
| 13 | 80ab308dfeat: Add Claude Plugin packaging and beta marketplace | Kaushalt2004 | 3w ago | +1.5K -224 | 89%Elevated | more lines added than baseline |
| 14 | b7d4c52dchore: downgrade version to 0.9.1, add benchmark report, and rewrite README | Kaushalt2004 | 1mo ago | +1.3K -232 | 89%Elevated | more lines added than baseline |
| 15 | a37a9864feat: NEXUS multi-agent runtime — agent registry, coordination bus, token tracker, trajectory store, 4 routing policies, 80 trajectories stored for offline RL | Kaushalt2004 | 2mo ago | +1.2K -21 | 89%Elevated | more lines added than baseline |
| 16 | 70445549feat: SWE-bench-lite integration, persistent SQLite cognition store, prompt mutation engine, 6/6 SWE tasks solved (+100% vs baseline) | Kaushalt2004 | 2mo ago | +993 -0 | 89%Elevated | more lines added than baseline |
| 17 | c6827391feat: CogniCore RealBench framework | Kaushalt2004 | 2mo ago | +1.0K -0 | 89%Elevated | more lines added than baseline |
| 18 | d0247c49feat: Add CogniBench benchmark suite UI | Kaushalt2004 | 2mo ago | +789 -0 | 89%Elevated | more lines added than baseline |
| 19 | 56bbabc0feat: Add MultiAgent, NPCSimulation, WorkflowAgent environments (62 total envs) | Kaushalt2004 | 2mo ago | +756 -0 | 89%Elevated | more lines added than baseline |
| 20 | 1b0aee11feat: 5 wow features - CognitiveBoost, Arena (ELO), AutoCurriculum, TransferLearning, MegaDemo | Kaushalt2004 | 2mo ago | +1.0K -0 | 89%Elevated | more lines added than baseline |
| 21 | d432dcb6rebuild: proper CLI terminals, Live Demo with Gemini API key, fixed title | Kaushalt2004 | 3mo ago | +1.1K -1.3K | 89%Elevated | more lines added than baseline |
| 22 | 4e385cb3Premium README + Vercel deployment setup (serverless API + dark-mode dashboard) | Kaushalt2004 | 3mo ago | +1.3K -136 | 89%Elevated | more lines added than baseline |
| 23 | 0d1ad75afix: inference.py uses HTTP calls, add root endpoint, baseline scores in README | Kaushalt2004 | 4mo ago | +1.5K -91 | 89%Elevated | more lines added than baseline |
| 24 | fbd6f374feat: CogniCore AI Safety Monitor v1.0.0 - AISafetyEnv with step/reset/state OpenEnv spec - 30 hand-crafted safety cases (easy/medium/hard) - Pydantic typed models (Observation, Action, Reward) - CogniCore memory + reflection + safety middleware - FastAPI server for HuggingFace Spaces - inference.py with exact [START][STEP][END] format - Dockerfile for containerized deployment | Kaushalt2004 | 4mo ago | +2.0K -0 | 89%Elevated | more lines added than baseline |
| 25 | cef9a891feat: research-grade benchmark — AST similarity, experiment tracking, multi-LLM, patch history, 10 bugs, 5/6 proven | Kaushalt2004 | 2mo ago | +784 -0 | 84%Elevated | more lines added than baseline |
| 26 | 4afd1544feat: CogniCore Runtime - universal cognition layer for any AI agent (6/6 tests pass) | Kaushalt2004 | 2mo ago | +989 -3 | 84%Elevated | more lines added than baseline |
| 27 | 0db43cc6feat: Add live agent demo to docs/ for GitHub Pages hosting | Kaushalt2004 | 2mo ago | +790 -649 | 84%Elevated | more lines added than baseline |
| 28 | c2e179b2v0.7.0: embedding memory, HuggingFace integration, benchmark leaderboard, README rewrite, CognitiveGymWrapper | Kaushalt2004 | 2mo ago | +1.2K -178 | 84%Elevated | more lines added than baseline |
| 29 | 54a798f6feat: world-class envs - MazeRunner, Trading, Survival (50 envs total, NOT a prompt optimizer) | Kaushalt2004 | 2mo ago | +946 -1 | 84%Elevated | more lines added than baseline |
| 30 | fb00a409add docs/ for GitHub Pages hosting | Kaushalt2004 | 3mo ago | +714 -0 | 84%Elevated | more lines added than baseline |
| 31 | 83d83291feat: dashboard UI with Overview, Simulate, API Test tabs — served by FastAPI | Kaushalt2004 | 4mo ago | +768 -928 | 84%Elevated | more lines added than baseline |
| 32 | 6aa3886ffeat: CodeRepairBench + Technical Evaluation Report (real bugs, real tests, real metrics) | Kaushalt2004 | 2mo ago | +580 -0 | 80%Elevated | more lines added than baseline |
| 33 | 4c7d67dbfeat: cognition proof benchmark 5/5 proven | Kaushalt2004 | 2mo ago | +614 -0 | 80%Elevated | more lines added than baseline |
| 34 | e4d4a84dfeat: pygame renderer, TensorBoard logger, CLI tool, BattleArena env, GitHub Actions CI (8 gym envs, all pass check_env) | Kaushalt2004 | 2mo ago | +715 -1.3K | 80%Elevated | more lines added than baseline |
| 35 | 917cbb01feat: Gymnasium-native API + SB3 integration - PPO/DQN train on CogniCore envs (check_env PASSED) | Kaushalt2004 | 2mo ago | +726 -0 | 80%Elevated | more lines added than baseline |
| 36 | b00213b6feat: v0.6.0 - REAL RL: GridWorld, ResourceGathering, Q-Learning, SARSA, Genetic, Bandit agents | Kaushalt2004 | 2mo ago | +932 -3 | 80%Elevated | more lines added than baseline |
| 37 | 011f26effeat: v0.5.0 - real-world safety + code bugs environments (30 total envs) | Kaushalt2004 | 3mo ago | +723 -9 | 80%Elevated | more lines added than baseline |
| 38 | 05c29e98feat: add real executable demo scripts for conference recording | Kaushalt2004 | 1w ago | +416 -0 | 76%Elevated | more lines added than baseline |
| 39 | 418b0813feat: CodeRepairBench v2 - semantic rejection, strategy mutation, 5/5 proven (8/8 bugs solved vs 7/8 baseline) | Kaushalt2004 | 2mo ago | +552 -0 | 76%Elevated | more lines added than baseline |
| 40 | e181355dfeat: Add Universal Agent Evaluation Suite UI | Kaushalt2004 | 2mo ago | +483 -0 | 76%Elevated | more lines added than baseline |
| 41 | 7be5e9b2feat: Add developer training script - shows cognitive agent growth across 9 environments | Kaushalt2004 | 2mo ago | +473 -0 | 76%Elevated | more lines added than baseline |
| 42 | d8f22a54feat: multi-agent benchmark suite - tests 6 agent types across 4 envs (zero LLMs) | Kaushalt2004 | 2mo ago | +461 -0 | 76%Elevated | more lines added than baseline |
| 43 | 71e81373fix: resolve openenv validate failures (uv.lock, pyproject.toml, server/app.py) | Kaushalt2004 | 3mo ago | +946 -2 | 76%Elevated | more lines added than baseline |
| 44 | f5297fb4feat: Context Preservation System + test fixes (Checkpoint 174) | Kaushalt2004 | 1w ago | +635 -29 | 72%Elevated | more lines added than baseline |
| 45 | 47d318a7feat: SWE-bench runner — 20/20 (100%) resolve rate on mini-bench | Kaushalt2004 | 2mo ago | +561 -3 | 72%Elevated | more lines added than baseline |
| 46 | 053c8d92feat: RL policy learner (minimal wins 89%), ablation suite (4 configs x 3 seeds), 30 new tasks (50 total), 460+ trajectories | Kaushalt2004 | 2mo ago | +519 -0 | 72%Elevated | more lines added than baseline |
| 47 | b879d1d9feat: 24 SWE-bench tasks, ablation study, multi-seed stats — 18.2/20 solve rate, 84.5% repeat reduction, 6/6 PROVEN | Kaushalt2004 | 2mo ago | +483 -478 | 72%Elevated | more lines added than baseline |
| 48 | 319f6232feat: CodeRepairBench v3 - real Gemini LLM integration, rate limit handling, semantic patch rejection, prompt mutation | Kaushalt2004 | 2mo ago | +409 -0 | 72%Elevated | more lines added than baseline |
| 49 | 74f542e8feat: Full cognitive dashboard with memory, reflection, heatmap, phase indicators | Kaushalt2004 | 2mo ago | +359 -354 | 72%Elevated | more lines added than baseline |
| 50 | f2b3f1ebfeat: v0.4.0 - print->logging (31 files), AgentProtocol, full type safety | Kaushalt2004 | 3mo ago | +454 -252 | 72%Elevated | more lines added than baseline |
Two views of the same model: where the cuts fall, and what commit shape lands you above them.
Every scored commit, binned on the raw 0 to 10 score rather than the percentile. Percentile ranks are uniform by construction, so that axis has no shape to draw. The dashed lines are the tercile cuts behind each row's priority pill.
The 168 most recent commits, on their own recency sample rather than the feed above: that defaults to risk-sorted, so reusing it would plot only the top tercile and call it the spread. Big and scattered is what the model penalises. Click a dot to open it.
cognicore-dev/cognicore-my-openenv has 187 commits in its history from 6 contributors, the first of them Apr 5, 2026. In the last 90 days 334 files were touched, 482 times in total, most often remote.py. Every commit is scored for change risk from its size, spread and the history of the files it touches.
