caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Project memory, focused development and business workflows, and evidence-gated skill learning.
Plugin erol from marketplace erol in erolemir/EROL. Commands read from the repo's manifest — 1 month ago.
accessibility-auditImprove interactive UI access through keyboard, semantics, focus, and perceivable feedback.
analytics-event-contractsDefine analytics events with stable identities, semantics and privacy-aware verification.
api-contract-designEvolve service contracts with compatibility, validation, and failure semantics.
architecture-decisionMake bounded architectural decisions backed by repository constraints and explicit tradeoffs.
authentication-session-securityReview login, tokens and session lifecycle for binding, expiry, revocation and recovery weaknesses.
authorized-security-testingPlan and execute bounded security tests on explicitly authorized targets with reproducible findings.
background-job-reliabilityMake queued work recoverable through acknowledgements, leases, bounded retries and safe replay.
brand-positioningDevelop positioning from customer alternatives, credible differentiation and proof.
caching-strategyDesign cache keys, invalidation and freshness policies from measured read behavior and correctness requirements.
ci-failure-triageDiagnose build and CI failures through reproducible environment comparisons.
cloud-cost-optimizationReduce cloud spend using attributable usage, unit costs and reversible capacity decisions.
configuration-managementMake configuration explicit, validated and environment-consistent without exposing secrets.
container-runtime-hardeningReduce container privileges and image exposure while preserving required runtime behavior.
content-distribution-planPlan reusable content and channel distribution around audience needs, capacity and measurable outcomes.
context-handoffPrepare bounded task and agent handoffs with evidence and explicit unresolved work.
conversion-funnel-analysisAnalyze conversion funnels with consistent identities, event ordering and meaningful denominators.
customer-discoveryPlan unbiased customer discovery that separates observed problems from reactions to a proposed solution.
customer-segmentationDefine actionable customer segments from needs, behavior and reachable evidence rather than invented personas.
dashboard-designDesign operational dashboards around decisions, metric definitions and comparable data states.
data-pipeline-integrityBuild batch or streaming pipelines with explicit schemas, reconciliation and recoverable checkpoints.
database-backup-restoreDesign backup and restore procedures around recoverable data, dependencies and measured recovery objectives.
database-migration-safetyPrepare compatible schema changes with recovery and data-integrity verification.
database-query-performanceMeasure and improve slow SQL through query plans and representative workload evidence.
dependency-boundariesRestructure modules around dependency direction, ownership and testable interfaces.
dependency-supply-chain-reviewReview dependency provenance, build scripts and artifact integrity for supply-chain exposure.
dependency-upgradeUpgrade dependencies through compatibility research, reproducible resolution, and rollback evidence.
design-system-componentsCreate reusable UI components with explicit variants, interaction contracts and consistent tokens.
disaster-recoveryPlan disaster recovery across infrastructure, data, credentials and routing with a tested failback boundary.
distributed-workflowCoordinate multi-service effects using durable states, reconciliation and explicit compensation limits.
documentation-maintenanceUpdate project documentation from observed behavior and reproducible examples.
error-handling-contractsDesign errors that preserve causality, stable caller behavior and safe operational diagnostics.
evidence-code-reviewReview changes for concrete correctness risks with reproducible findings.
experiment-statisticsEvaluate experiments with valid randomization, uncertainty and predeclared outcome decisions.
file-upload-pipelineBuild resumable file ingestion with bounded resources, durable ownership and visible processing states.
form-validationImplement form validation, submission and recovery without losing user input or trusting client checks alone.
frontend-data-fetchingManage remote UI data through cache keys, cancellation and consistent loading/error states.
frontend-e2e-testingTest critical browser journeys with stable state setup, meaningful assertions and controlled dependencies.
frontend-state-correctnessDiagnose stale UI, asynchronous races, and inconsistent client state transitions.
git-branch-integrationSynchronize and integrate Git branches while preserving local work and divergent history.
growth-experiment-designDesign bounded business growth experiments with falsifiable hypotheses, guardrails and decision rules.
incident-debuggingInvestigate reproducible failures and preserve verified causes and regression evidence.
infrastructure-change-planReview infrastructure changes for drift, state ownership, blast radius and recoverable application impact.
internationalizationImplement locale-aware UI strings, formats and directionality without changing canonical stored values.
kubernetes-rolloutPlan recoverable Kubernetes deployments using readiness, capacity, versioned artifacts and rollback checks.
landing-page-optimizationImprove landing pages through message/offer alignment, user friction and measured conversion hypotheses.
learning-pattern-discoveryTurn repeated verified incidents into bounded project workflow candidates.
legacy-modernizationModernize legacy code incrementally with characterization, compatibility seams and recoverable rollout.
lifecycle-email-marketingDesign permission-aware lifecycle emails triggered by meaningful product states and measured user outcomes.
linux-service-operationsDiagnose Linux service health from process ownership, logs, resources and configuration before changing service state.
llm-feature-evaluationEvaluate LLM features on representative fixtures, failure categories, cost and human-review requirements.
market-researchResearch a market and competitors with dated evidence, explicit segments and defensible comparisons.
marketing-copywritingWrite truthful campaign or product copy matched to audience intent, channel constraints and a clear action.
memory-hygienePreserve small, evidence-backed project memory with redaction, staleness, and provenance.
message-idempotencyPrevent duplicate side effects in retried requests and at-least-once message consumers.
multi-tenant-isolationVerify tenant scoping across queries, jobs, caches and object references.
observability-designDesign logs, metrics and traces that explain user-visible failures without excessive cardinality or sensitive payloads.
paid-search-campaignDesign paid-search campaign drafts with intent separation, budget boundaries and reliable conversion measurement.
paid-social-campaignPrepare paid-social creative tests with credible messages, audience boundaries and budget-aware measurement.
partnership-strategyEvaluate partnerships from mutual customer value, economics, responsibilities and pilot evidence.
performance-profilingMeasure bottlenecks before optimizing CPU, memory, or end-to-end latency.
pricing-packaging-researchResearch product pricing and packaging from customer value, unit economics and defensible willingness-to-pay evidence.
privacy-data-retentionMap sensitive data collection, retention and deletion to concrete systems and stated policy requirements.
product-prioritizationPrioritize product work using customer evidence, constraints and transparent impact assumptions.
production-incident-responseRespond to production incidents with evidence preservation, scoped mitigation and recovery verification.
rate-limit-designDesign bounded request quotas with explicit identities, burst behavior and fair failure responses.
realtime-serviceDesign authenticated realtime connections with bounded delivery, reconnect and backpressure behavior.
regression-test-designSelect focused tests that demonstrate behavioral changes and protect meaningful boundaries.
release-verificationPrepare evidence-backed releases with explicit readiness gates and recoverable rollout.
repository-change-planningPlan a scoped code change from affected callers, compatibility constraints and verifiable milestones.
repository-explorationFind relevant code paths and project conventions before planning a change.
requirements-acceptanceTurn feature requests into observable acceptance criteria, boundaries and unresolved decisions.
responsive-layoutBuild layouts that remain usable across viewport sizes, zoom and variable content.
retention-cohort-analysisMeasure retention using comparable cohorts, meaningful activity definitions and censoring-aware windows.
reverse-proxy-tlsConfigure proxy routing and TLS while preserving trusted client identity and upstream request semantics.
safe-refactoringChange structure while preserving externally observable behavior and project style.
sales-pipeline-designDesign a sales pipeline around buyer evidence, stage exit criteria and reliable operational reporting.
secrets-rotationPlan and verify credential rotation with overlap, revocation and recoverable dependent-service updates.
secure-file-handlingReview untrusted file parsing, paths and storage boundaries for traversal and resource exhaustion.
security-boundary-reviewExamine untrusted inputs, authorization boundaries, secret handling, and dangerous side effects.
security-log-detectionDesign actionable security detections from reliable event semantics and tested false-positive controls.
seo-content-briefseo-international-sitesseo-migrationseo-performance-measurementseo-structured-dataseo-technical-auditservice-api-integrationskill-quality-evaluationstable-paginationtest-fixture-designthreat-modelingtool-capability-selectiontransaction-boundariesvulnerability-triageweb-performance-budgetwebhook-deliveryerolTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.