cyberneticlibrary

The library

Everything we index — ranked by what works, never by stars.

WORKS 55
Resolve all parity findingsworkflowEngineeringL3
parity-fix · Iterative multi-phase workflows with safeguards.
WORKS 56
Audit entire codebase for bugsworkflowEngineeringL3
compass-audit · A deep audit of an area or the whole repo — more coverage than one review pass. Scope it with args.path (e.g.
WORKS 55
Evaluate sprint contract dynamicallyworkflowEngineeringProductL3
harness-eval · Iterative multi-phase workflows with safeguards.
WORKS 55
Audit dependencies for securityworkflowEngineeringOpsL3
dependency-audit · Monthly per repo, before major releases, or when adding a significant new dependency. Surface critical security findings
WORKS 54
Compare AI models dailyworkflowDataProductL3
model-arena-daily · Daily intelligence on which Claude tier is best for each task type. Cost-disciplined by design: 3 generators + 1 judge =
WORKS 55
Verified swarm executionworkflowL2
verified-swarm.workflow · Catching error-suppression in automated fixes via voting.
WORKS 60
Decompose issue into sub-issuesworkflowEngineeringProductL2
decompose-issue · When a task is too large for one ticket and should become a parent + several sub-issues on phaabe/live.moafunk.de
WORKS 58
Review code diff for conventionsworkflowEngineeringL3
review-diff · Catching error-suppression in automated fixes via voting.
WORKS 55
Prepare dispatch scaffoldingworkflowEngineeringProductL2
dispatch-prep · When Leader wants to dispatch the Developer subagent manually in main chat but still benefit from the preflight + scaffo
WORKS 55
Sweep for untested business logicworkflowEngineeringFinanceL3
coverage-sweep · Iterative multi-phase workflows with safeguards.
WORKS 55
Verify commit claims before pushworkflowEngineeringL2
pre-push-verify · Before git push of a feature branch, before gh pr create, before gh pr merge. Especially after rebase / amend / worktree
WORKS 55
Draft feature flags in parallelworkflowEngineeringL3
ws-bf-fanout · Catching error-suppression in automated fixes via voting.
WORKS 53
Discover targets through iterative deduplicationworkflowSalesMarketingL3
loop-discover · When the amount of work is unknown upfront — bugs, proposals, patterns. Spawns finder agents until K consecutive rounds
WORKS 55
Generate verified science curriculum questionsworkflowProductEngineeringL3
author-content · Generating science curriculum MCQ items with guaranteed content accuracy and grade-fit validation through adversarial verification.
WORKS 58
Batch-process design chapter rewritesworkflowProductL3
author-rework-batch · Iterative multi-phase workflows with safeguards.
WORKS 54
Audit AI game engine capability gapsworkflowEngineeringL3
p6-complete-dm-gap-audit · Iterative multi-phase workflows with safeguards.
WORKS 58
Polish workflow visualization mapsworkflowEngineeringL3
polish-ascii-map · Iterative multi-phase workflows with safeguards.
WORKS 56
Improve workflow GUI usabilityworkflowProductEngineeringL3
polish-gui · Iterative multi-phase workflows with safeguards.
WORKS 55
Multi-agent Go code review pipelineworkflowEngineeringL3
verify-go-review · Catching error-suppression in automated fixes via voting.
WORKS 57
Parallelize sprint story implementationworkflowEngineeringL3
harness-implement-group · Iterative multi-phase workflows with safeguards.
WORKS 55
Diagnose agent evaluation failuresworkflowEngineeringL3
wave-analysis · Root-cause analysis across multi-scenario runs with dedup and regression detection.
WORKS 57
Run security scans across OWASP Top 10workflowEngineeringLegalL3
forge-check-security · OWASP parallel scanning (6 dimensions) with risk aggregation and gate enforcement.
WORKS 58
Audit research findings for completenessworkflowDataL3
grounding-audit · Parallel read-only audits against design claims with completeness criticism.
WORKS 60
Fix and test path traversal vulnerabilityworkflowEngineeringL3
fix-harvest-security · OWASP parallel scanning (6 dimensions) with risk aggregation and gate enforcement.
WORKS 55
Generate daily research briefworkflowDataMarketingL3
research-pulse-daily · Daily lightweight research pulse: 3 parallel domain scans + synthesis.
WORKS 58
Harden committer resource injectionworkflowEngineeringL3
fix-committer-harvest · Security hardening: fixer → test-writer → adversarial reviewer loop.
WORKS 54
Implement feature with review loopworkflowEngineeringL2
implement-and-review · Iterative code review → fix loops up to 3 rounds, hard-capped.
WORKS 55
Compare options with scored tableworkflowProductOpsL2
compare-options · Head-to-head option comparison: parallel research + impartial scoring.
WORKS 58
Compose parallel review verificationworkflowEngineeringL3
trace-compose · Test multi-stage pipeline with nested parallel (canonical review→verify shape).
WORKS 55
Workflow orchestration toolworkflowL2
WorkflowTool · JavaScript workflow orchestration engine with resume/caching support.
WORKS 54
Certify specification complianceworkflowEngineeringL3
skv18-s-p3-certify · Multi-lens schema certification with falsifiability and sequencing audits.
WORKS 59
Deep Python file security reviewworkflowEngineeringL3
file-review · Code review over file set: parallel analysis + consensus.
WORKS 54
Parallelize panic diagnosisworkflowEngineeringL3
phase-c-panic-swarm · Crisis response: parallel threat detection + rapid escalation.
WORKS 57
Implement and verify keystoneworkflowEngineeringL3
phase-b2-keystone · Critical path validation: parallel checks on load-bearing components.
WORKS 55
Parallelize complex coding slicesworkflowEngineeringL3
coding · Code generation: prompt → parallel agents → merge.
WORKS 55
Validate skill documentation driftworkflowEngineeringOpsL3
audit-skill-consistency · Skill consistency audit across agents/dimensions.
WORKS 54
Classify memory safety issuesworkflowEngineeringL2
phase-h-classify-issues · Prioritizing crash/memory-safety issues in a Zig→Rust migration by classifying GitHub issues at scale with Rust-port impact assessment.
WORKS 55
Run code review debates until consensusworkflowEngineeringL3
codex-debate · Adversarial debate to uncover contradictions in claims/code.
WORKS 48
Build company research source catalogworkflowDataL3
driver-menu-build · Interactive menu-driven workflows with dynamic branching.
WORKS 60
Gate releases with automated quality checksworkflowEngineeringL3
release-gate · Release gating with multi-criteria pass/fail verdicts.
WORKS 54
Implement and test single issuesworkflowEngineeringL2
close-issue · Issue closure with verification and state logging.
WORKS 54
Deploy hotfixes with minimal reviewworkflowEngineeringL2
hotfix · Orchestrate multi-phase hotfix tasks with structured verification and recovery.
WORKS 58
Build MoE-aware model distillationworkflowEngineeringL3
frontier-tail-build-w2-moe · Orchestrate multi-phase frontier-tail-build-w2-moe tasks with structured verification and recovery.
WORKS 54
Verify findings across parallel reviewersworkflowEngineeringL3
review-changes · Orchestrate multi-phase review-changes tasks with structured verification and recovery.
WORKS 55
Generate DSP reference documentationworkflowEngineeringL3
ugen-dsp-notes · Orchestrate multi-phase ugen-dsp-notes tasks with structured verification and recovery.
WORKS 56
Assign game palettes to level stagesworkflowProductL2
rpg-stages · Orchestrate multi-phase rpg-stages tasks with structured verification and recovery.
WORKS 55
Explore RPG redesign directionsworkflowProductL2
rpg-design · Orchestrate multi-phase rpg-design tasks with structured verification and recovery.
WORKS 57
Optimize command descriptions for reliabilityworkflowEngineeringL3
description-optimize · Orchestrate multi-phase description-optimize tasks with structured verification and recovery.
WORKS 60
Scale model training with grid searchworkflowEngineeringL3
dead-presidents-resweep-scale · Orchestrate multi-phase dead-presidents-resweep-scale tasks with structured verification and recovery.
WORKS 57
Autopilot feature requests to executionworkflowProductEngineeringL3
auto-feature · Orchestrate multi-phase auto-feature tasks with structured verification and recovery.
page 21 / 156