cyberneticlibrary

The library

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

WORKS 48
Discover and fill knowledge gapsworkflowProductEngineeringL2
mishkan-knowledge-gap-discovery · When auditing whether expected knowledge exists or needs research to fill gaps.
WORKS 48
Review and cluster product backlogworkflowProductL2
lets-backlog · When domain experts need to surface backlog gaps and cluster related themes in parallel.
WORKS 60
Execute event pipeline end-to-endworkflowOpsL3
sweep-civ · When batch-upgrading timeline events with photos and verification across multiple gates.
WORKS 48
Conduct adversarial security threat auditworkflowOpsL3
threat-audit · When identifying threat vectors requires multiple independent adversarial security lenses.
WORKS 48
Scale agents within token budgetworkflowEngineeringOpsL2
loop-until-budget · When a task needs to run repetitively until completion or token budget exhaustion.
WORKS 48
Run complete CI quality gatesworkflowEngineeringL3
full-quality-gate · When integrating code and docs require automated compliance gates before promotion.
WORKS 48
Design and execute test strategyworkflowEngineeringProductL2
test · When running tests needs detailed instrumentation and result aggregation.
WORKS 48
Validate architecture and acceptance criteriaworkflowEngineeringL3
kiln-validate · When validating codegen artifacts must match strict schema conformance rules.
WORKS 48
Optimize with evolutionary tree searchworkflowEngineeringProductL4
evo-optimize · When iteratively refining codegen output requires evolutionary feedback loops.
WORKS 48
Fan-out multi-agent dispatchworkflowEngineeringL2
fan-out.template · When executing parallel agents on the same arguments without manual orchestration.
WORKS 48
Audit cryptographic vault securityworkflowEngineeringLegalL3
env-ctl-audit-phase1 · When auditing infrastructure configuration state against policy in parallel.
WORKS 48
Verify implementation with skepticsworkflowEngineeringProductL3
adversarial-verify · When testing correctness claims requires independent skeptics voting on evidence.
WORKS 48
Execute software spec end-to-endworkflowEngineeringL3
software-dev-pipeline · When driving full software development with design, implementation, verification, and deployment.
WORKS 48
Summarize Jira sprint digestworkflowOpsEngineeringL2
jira-digest · When generating team digests from Jira activity with structured, parseable output.
WORKS 55
Multi-lens codebase auditworkflowEngineeringL3
mishkan-codebase-audit · Periodic project-wide audits, pre-release reviews, post-incident hardening passes.
WORKS 48
Verify spec against requirementsworkflowProductEngineeringL3
opsx-verify · Run after apply to prove each spec requirement is satisfied.
WORKS 55
Run parallel code reviewsworkflowEngineeringL3
parallel-review · Use when workflow operations need parallel review.
WORKS 55
Four-wave deterministic code reviewworkflowEngineeringL3
comprehensive-review-workflow · Use when workflow operations need comprehensive review workflow.
WORKS 48
Map grant field coverage gapsworkflowFinanceLegalL3
field-coverage · Use when workflow operations need field coverage.
WORKS 48
Draft and score implementation plansworkflowProductEngineeringL3
compass-plan · Before a non-trivial or ambiguous change where the approach is not obvious.
WORKS 48
Discover, research, and designworkflowProductEngineeringL3
design · Use when workflow operations need design.
WORKS 48
Verify workshop slides and diagramsworkflowMarketingOpsL3
workshop-verify · Called by the workshop skill at the Phase 3->4 boundary (artifact review gate) and as Phase 4 verification, and by workshop-revise after edits.
WORKS 55
Parallel file mutations with isolationworkflowEngineeringL3
worktree-isolation · Use when workflow operations need worktree isolation.
WORKS 48
Refactor transformer block with gatesworkflowEngineeringL3
finish-l2-transformer-block · Continuing the toy-framework P2 refactor after L1: extract the L2 TransformerBlock as a single gated unit.
WORKS 48
Autonomous ML research loopworkflowDataEngineeringL3
{PROJECT_NAME}-autoresearch · Use when workflow operations need {PROJECT_NAME} autoresearch.
WORKS 48
Extract and merge literature reviewworkflowDataProductL3
paper-review-literature · Use when workflow operations need paper review literature.
WORKS 48
Audit cross-chain configuration rulesworkflowEngineeringProductL3
long-chain-diagnostics-audit · Use when workflow operations need long chain diagnostics audit.
WORKS 48
Adversarially review research paperworkflowDataOpsL3
review-paper · Use when workflow operations need review paper.
WORKS 48
Classify and synthesize memory entriesworkflowOpsProductivityL3
memory-gc · Use when workflow operations need memory gc.
WORKS 48
Triple-LLM parallel code reviewworkflowEngineeringL3
cr-multi · Use when workflow operations need cr multi.
WORKS 48
Map tools vs effectTools across SDKsworkflowEngineeringProductL3
tools-vs-effecttools · Use when workflow operations need tools vs effecttools.
WORKS 48
Orchestrate tiered task executionworkflowEngineeringOpsL3
wfl:run-task · Run a single Forge task through its full plan→implement→review→approve→commit pipeline via a deterministic JS driver instead of the LLM orchestrator.
WORKS 48
Fan-out deep research with citationsworkflowDataMarketingL3
deep-research · Use when workflow operations need deep research.
WORKS 48
Dispatch bound doc reviewersworkflowOpsEngineeringL3
oagp-seat-doc-reviewer · Use when workflow operations need oagp seat doc reviewer.
WORKS 48
Plan implementation slices with orchestrationworkflowEngineeringL2
fathomdb-0.8.0-slice-plan · Planning large multi-phase database implementations with complex dependency sequencing and human-in-loop gates.
WORKS 58
Complete WarmStart and Curriculum recipesworkflowEngineeringL3
finish-l4-warmstart-curriculum · Building deterministic multi-stage training recipes with bit-identical gating vs. reference curves.
WORKS 48
Process PR comments and request reviewsworkflowEngineeringL2
comms-workflow · Continuous monitoring and triage of PR comments with automated fixes and intelligent re-request logic.
WORKS 48
Propagate design tokens across componentsworkflowEngineeringProductL3
panim-ds-rollout · Propagating design token changes across component consumers with a11y and visual regression gates.
WORKS 48
Research and verify specificity audit methodologyworkflowDataL3
step3-specificity-methodology · Auditing bispecific therapeutic window via parallel research, synthesis, and adversarial critique.
WORKS 53
Generate release notes from git historyworkflowEngineeringL2
sefer-release-notes · Assembling release notes from commits when different categories need orthogonal specialist perspectives.
WORKS 48
Ingest and index large books in parallelworkflowDataL3
ingest-mcluhan-understanding-media · Ingesting large multi-chapter books with deduplication and cross-chunk synthesis via chunked planners.
WORKS 48
Run parallel UI tests with PlaywrightworkflowEngineeringL2
playwright-parallel-test · Running parallel UI tests across form/nav/layout when isolation and concurrent feedback matter.
WORKS 48
Evolve multi-agent game-playing harnessworkflowEngineeringL3
codeclash-decomp-evolution · Evolving multi-agent harnesses via selection pressure with verified fitness gates and ablation analysis.
WORKS 48
Cross-check research claims with sourcesworkflowDataL2
deep-research · Running multi-source research with web fetching, fact verification, and cited synthesis.
WORKS 60
Dispatch and monitor pipeline executionworkflowOpsL3
software-factory-dispatcher · Orchestrating work across specialized factory agents with parallel task distribution.
WORKS 48
Extract CLAUDE.md candidates from sessionsworkflowEngineeringL2
mine-claude-md-from-sessions · Extracting and synthesizing codebase documentation from session transcripts.
WORKS 48
Execute full Loom retrieval pipelineworkflowEngineeringL3
loom-orchestrate · Coordinating loom weaving with pattern composition and constraint satisfaction.
WORKS 48
Generate files from design specificationworkflowEngineeringL3
wc-generate · Generating contextual content via word-chain models and sampling.
WORKS 48
Execute planner-executor-verifier loopworkflowOpsL3
traveleditor-stream · Streaming editorial changes in a travel editor with real-time sync.
WORKS 48
Generate project context snapshotworkflowEngineeringL2
context-refresh · Refreshing and updating cached context for downstream tasks.
page 18 / 24