Free forever · MIT licensed

Free Claude Skills for Developers

Token-lean, secure-by-default skills for Claude Code and agentic workflows. Free to download, MIT-licensed.

45 skills · 1 system · 19 categories · free forever · 17 downloads

Systems

· complete products, beyond single skills

SystemFree · MIT

SYNAPSE — Perpetual Memory & Shared Brain

synapse · v3.5

A complete memory system for AI coding agents in a single markdown file. Your agent never forgets.

memorybrainpersistenceclaude-code0

SYNAPSE — Perpetual Memory & Shared Brain

Explore

AI / LLM

· 3

Free

RAG Designer

rag-designer · v1.0.0

End-to-end RAG that returns grounded answers — semantic chunking, hybrid retrieval + re-rank, grounded generation, separate eval. Retrieval-first. A comprehensive skill.

ragretrievalembeddings

~969 tokens0

Free

LLM Eval Designer

llm-eval-designer · v1.0.0

Turn 'it feels better' into a number — real varied cases, programmatic-first judging, per-dimension baseline, regression suite for a non-deterministic system.

evalllmtesting

~752 tokens0

Free

Prompt Engineer

prompt-engineer · v1.0.0

Reliable LLM output from structure + examples + guardrails, not magic words — with an explicit out for uncertainty. Iterate on evidence.

prompt-engineeringllmfew-shot

~713 tokens0

API Design

· 2

Free

REST API Designer

rest-api-designer · v1.0.0

Consistent, predictable REST — noun resources, right status codes, one error shape, cursor pagination, versioned. Boring beats clever.

restapihttp

~683 tokens0

Free

OpenAPI Writer

openapi-writer · v1.0.0

A complete OpenAPI spec grounded in the real endpoints — every response status, reusable schemas, real examples. Not a half-spec.

openapiswaggerapi-docs

~538 tokens0

Code Quality

· 2

Free

Code Review

code-reviewer · v1.0.0

Two passes on a diff — correctness bugs first, real cleanups second, zero style nitpicks. Every blocker carries a concrete failure scenario.

code-reviewcorrectnesspull-request

~590 tokens0

Free

Simplify Code

simplify-code · v1.0.0

Cut dead code, over-abstraction and needless nesting — behavior preserved, prefer deletions. Knows what NOT to touch.

simplificationreadabilitydead-code

~468 tokens0

Content

· 1

Free

Newsletter Writer

newsletter-writer · v1.0.0

Inbox-first email that gets opened and read — one message, one CTA, 3 A/B subject variants, brand voice controls. Not SEO content.

newsletteremailcopywriting

~501 tokens0

Data

· 2

Free

Data Cleaner

data-cleaner · v1.0.0

Clean a dataset without silently corrupting it — profile first, log every transform, keep the original, reconcile row counts.

datacleaningvalidation

~629 tokens0

Free

Notebook Structurer

notebook-structurer · v1.0.0

Make an analysis notebook reproducible — Restart & Run All clean, narrative order, no hidden state, dead cells gone.

jupyternotebookreproducibility

~610 tokens0

Database

· 2

Free

DB Migration Auditor

db-migration-auditor · v1.0.0

Catch the outage on paper: data loss, table locks, breaking renames, missing RLS — before the migration runs. Postgres/Supabase-aware.

migrationddlpostgres

~675 tokens8

Free

SQL Query Optimizer

sql-query-optimizer · v1.0.0

EXPLAIN-driven, not folklore: read the plan, fix the expensive node, verify it moved. Recommends only indexes the planner will use.

sqlperformanceexplain

~587 tokens0

Dev Workflow

· 5

Free

Guardrails (Token-Lean)

guardrails-token-lean · v1.0.0

Estimate the token bill before you spend it — right-size fan-out, model tier, and reads to the task, then stop once the returns flatten.

token-efficiencysafetybehavioral

~731 tokens0

Free

Plan Interrogator

plan-interrogator · v1.0.0

Stress-test any plan in max 3 bounded rounds — exit with a build-ready spec, not an endless interview.

planningspecrequirements

~839 tokens0

Free

Refactor (Behavior-Preserving)

refactor-behavior-preserving · v1.0.0

Restructure code without changing behavior, proven green by tests, scoped to what you touched. No bugfixes smuggled into a 'cleanup'.

refactoringsimplificationarchitecture

~545 tokens0

Free

Systematic Debugging

systematic-debugging · v1.0.0

Reproduce → isolate → prove the cause → fix the root. Triages trivial bugs fast, goes deep only when needed. Ships drop-in instrumentation.

debuggingroot-causetriage

~569 tokens0

Free

TDD Workflow

tdd-workflow · v1.0.0

Red-green-refactor with a gate that blocks code before its failing test exists — light enough for a single function.

tddtestingred-green-refactor

~608 tokens0

DevOps

· 3

Free

CI/CD Pipeline Builder

ci-pipeline-builder · v1.0.0

Fast feedback, safe deploys — fail-fast stage order, lockfile-keyed caching, no leaked secrets, gated progressive delivery.

ci-cdgithub-actionspipeline

~576 tokens0

Free

Dockerfile Optimizer

dockerfile-optimizer · v1.0.0

Smaller, faster, safer images — multi-stage, manifest-before-source caching, non-root, no secrets in layers. Verified still-runs.

dockerdockerfileimage-size

~558 tokens0

Free

Terraform Diagnostic

iac-terraform-diagnostic · v1.0.0

Read the plan before you apply — catch destroy/replace of stateful resources, drift, and state traps. Diagnostic-first, not doc-dumping.

terraformiacinfrastructure

~660 tokens0

Documentation

· 2

Free

Changelog Writer

changelog-writer · v1.0.0

Turn commits/PRs into human-first release notes — breaking changes first, internal noise dropped, semver sanity-checked.

changelogrelease-notessemver

~463 tokens0

Free

README Writer

readme-writer · v1.0.0

A README a stranger can follow to running in minutes — grounded in the real code, updated incrementally, zero invented commands.

readmedocumentationonboarding

~485 tokens0

Frontend

· 3

Free

Accessibility Audit

accessibility-audit · v1.0.0

WCAG audit that maps each barrier to a criterion and a concrete fix — keyboard, names, contrast, semantics, live regions.

accessibilitya11ywcag

~663 tokens0

Free

Component Architecture

component-architecture · v1.0.0

Right-size a component — one responsibility, least state, composition over 15 boolean props. Fix the boundary, not the symptom.

componentsarchitectureprops

~598 tokens0

Free

Responsive Layout Check

responsive-layout-check · v1.0.0

Catch the breakpoint failures before they ship — overflow, touch targets, media, layout shift — fixed at the layout level.

responsivemobilecss

~565 tokens0

Git

· 1

Free

Git Flow

git-flow · v1.0.0

Conventional commits, PR descriptions and worktrees — with a validation gate that runs before you write history.

commitsconventional-commitspull-requests

~881 tokens2

Learning

· 2

Free

Concept Explainer

concept-explainer · v1.0.0

Clear AND accurate — builds from what you know, one concrete anchor (and where it breaks), progressive depth, flags every simplification.

explainlearningeli5

~602 tokens0

Free

Explain Code

explain-code · v1.0.0

Purpose → flow → gotchas at the altitude you asked for — reads the real code, corrects misleading names, skips line-by-line noise.

explainonboardinglegacy-code

~543 tokens0

Meta / Skills

· 2

Free

MCP Builder

mcp-builder · v1.0.0

Scaffold an MCP server with the patterns generators miss — auth, pagination, real errors, least-privilege — and ship it with a test.

mcpservertools

~553 tokens0

Free

Skill Creator (Secure)

skill-creator-secure · v1.0.0

Write a focused, token-lean skill AND lint it for safety before shipping — the generator that doesn't produce vulnerable skills.

skill-creatormetasecurity-lint

~577 tokens0

Performance

· 2

Free

Frontend Perf Auditor

frontend-perf-auditor · v1.0.0

Measure the waterfall/bundle/profile, fix the biggest bottleneck, verify it moved. No optimizing on hunches.

performancefrontendbundle-size

~561 tokens0

Free

Backend Profiler

backend-profiler · v1.0.0

Find the one dominant cost — N+1, blocking I/O, hot path — with evidence, fix by leverage, verify under real load.

performancebackendn+1

~560 tokens0

Productivity

· 2

Free

Memory / Handoff

memory-handoff · v1.0.0

Zero-infra file memory the next session picks up cold — persists decisions, constraints and gotchas, not what the repo already records.

memoryhandoffcontext

~518 tokens0

Free

Env Doctor

env-doctor · v1.0.0

'Works on my machine', fixed: checks version → deps → env vars → ports → paths in order, and diffs against CI when local breaks but CI passes.

environmentsetupworks-on-my-machine

~578 tokens0

Project Management

· 2

Free

Decision Recorder (ADR)

decision-recorder · v1.0.0

Preserve WHY a decision was made and which alternatives were rejected — append-only ADRs so the reasoning survives the team.

adrarchitecturedecision

~625 tokens0

Free

Task Estimator

task-estimator · v1.0.0

Decompose until pieces are real, count the invisible work, give a range, name the unknown that makes it 3x. Estimate the work, not the deadline.

estimationplanningbreakdown

~657 tokens0

Research

· 3

Free

Deep Research

deep-research · v1.0.0

Scope → fan out → adversarially verify every load-bearing claim → cited synthesis with confidence and dissent. A comprehensive research harness.

researchfact-checkingsources

~908 tokens7

Free

Competitive Analysis

competitive-analysis · v1.0.0

Honest competitor teardown — includes where you LOSE, grounded in what rivals actually do, ends on the positioning gap.

competitivemarket-researchpositioning

~632 tokens0

Free

Source Summarizer

source-summarizer · v1.0.0

A summary you can trust instead of the original — faithful to the real claims and caveats, adds nothing, flattens nothing.

summarizefaithfuldistillation

~584 tokens0

Security

· 3

Free

Security Review (Code)

security-review-code · v1.0.0

Vulnerability review of a diff with executable sweeps + OWASP/CWE mapping. Findings ranked by demonstrated exploit, each with a concrete fix.

vulnerabilityowaspcwe

~599 tokens0

Free

Skill Security Auditor

skill-security-auditor · v1.0.0

Audit any third-party skill before you install it — prompt injection, exfiltration, dangerous commands. The check the marketplaces don't do.

supply-chainauditprompt-injection

~693 tokens0

Free

Secrets Guardian

secrets-guardian · v1.0.0

Stop API keys, tokens and private keys from entering code, commits or logs — detect, triage false positives, and fix the pattern.

secretscredentialspre-commit

~570 tokens0

Testing

· 3

Free

Flaky Test Detective

flaky-test-detective · v1.0.0

Kill intermittent test failures at the source — timing, shared state, clock/RNG, races — never with a retry band-aid.

flakytestingnon-determinism

~544 tokens0

Free

Test Generator

test-generator · v1.0.0

Tests that would actually catch a bug — derived from the contract, one behavior each, no coverage-gaming.

testingunit-testscoverage

~548 tokens0

Free

E2E Test Writer

e2e-test-writer · v1.0.0

End-to-end tests for the flows that matter — stable selectors, self-contained data, user-visible assertions.

e2eplaywrightcypress

~539 tokens0

Built by someone who ships

Every skill in this catalog is written and maintained by Anthony Martinez — token-lean, secure-by-default, and tested against real 2026 agent workflows before release. Read our editorial philosophy →