SuperSkill Registry MCP server
online · 96% uptimecom.onlyharness/registry·v0.3.1Find, inspect and install reviewed reusable AI-agent capabilities from superskill.sh.
0 stars
Reviews
Write oneNobody has reviewed SuperSkill Registry yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
SuperSkill Registry tools (12, 2 write)
write = sends, deletes, buys or postsfetchFreeFetch the complete public catalog record for an exact resource id returned by search, including a canonical citation URL.
harness_detailFreeReturn manifest, trust signals, example and file list for one harness.
publish_markdown_to_harnesswrite actionFreeConvert markdown into an unverified harness scaffold. Requires a SuperSkill Bearer token.
publish_resource_packagewrite actionFreePublish an immutable canonical hosted package release. Requires a confirmed-user Bearer with active superskill:managed scope, semantic version and idempotency key. Returns unreviewed trust; never grants a Verified badge.
pull_harnessFreeReturn archive files for a harness. Paid harnesses return payment requirements unless the Bearer token is entitled.
pull_instructionsFreeReturn CLI and HTTP commands for pulling a harness into a local workspace.
resource_detailFreeReturn provenance, trust, popularity and actions for one mixed resource. Pass version to bind a hosted immutable release; latest is never substituted for an explicit version.
resource_use_instructionsFreeReturn the best safe next action for a mixed resource. Pass version to bind an immutable hosted release. Hosted skills expose explicit-consent exact native client install commands; upstream-only resources stay open-only.
searchFreeSearch public agent resources. Returns exact resource ids and canonical browser-openable URLs for citation.
search_docsFreeSearch /llms.txt and agent guidance for API, CLI, MCP and safety instructions.
search_harnessesFreeSearch the SuperSkill registry by task, job, title, summary or tag.
search_resourcesFreeSearch mixed source-aware resources: harnesses, skills, plugins, workflows, MCP servers, configs, guides, runtimes and directories.
Public scan report
scanner v0.1.9 · 2026-09-24 · same rubric, same numbers if you re-run it
- –Code scanremote-only server, no package to scann/a
- Live reliabilityremote reachable in 1237ms20/20
- Tool poisoning12 tool descriptions checked15/15
- Auth qualityopen endpoint exposes 2 write-action tools with no auth3/15
- Maintenancelast push 55 days ago12/15
- Maintainer identitynamespace and repository owner differ; GitHub account older than a year5/10
Findings (1)
- highWrite-action tools reachable without authentication
auth.open-write
How SuperSkill Registry compares
Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 73/100
- higher than 42% of 4,460 AI and agents servers
- Tools
- 12
- the typical AI and agents server has 6
- Tools that change things
- 17%
- typical: 0%. Fewer write tools means less an agent can break
- Uptime
- 96%
- of our recent checks; the typical remote AI and agents server answers 100%
- Response time
- 155 ms
- faster than 76% of 2,331 AI and agents servers
What its tools let an agent do: 10 read · 2 send or post
Install SuperSkill Registry in Claude Code, Cursor or VS Code
claude mcp add --transport http registry https://superskill.sh/mcp
What the publisher says
From the SuperSkill Registry repository's README, as published. We do not edit it. Read it on GitHub
SuperSkill
superskill.sh is the single human-facing product for finding and installing reviewed AI-agent capabilities. It preserves existing registrations and workspaces, exposes one universal install link for Codex and Claude Code, and keeps exact release, trust, revocation and consent gates explicit.
The old OnlyHarness UI and host are compatibility-only: human routes redirect to SuperSkill, while old API/MCP coordinates, npm/package names, database identifiers and immutable archive URLs remain available for installed clients and data continuity. Internal package names still use @harnesshub/*; the published compatibility CLI package remains onlyharness.
What is a native harness package?
A native harness package is the strict, verified package format inside the broader OnlyHarness resource catalog:
- harness.yaml manifest with runtime, tools, permissions, quality gates, and risk profile.
- Prompt, examples, eval cases, and expected outputs.
- CLI commands for validate, eval, gate, diff, import, and PR annotation.
- Social layer: stars, server-side remix forks, threads, verified gate runs, install confirms, heat, tags, outcomes, and maintainer review.
Live MVP
- App: https://superskill.sh
- API health: https://superskill.sh/api/healthz
- Registry API: https://superskill.sh/api/registry
- Universal installer manifest: https://superskill.sh/api/superskill/install
Supabase auth is enabled for signup/login, stars, local remix drafts, thread posts, and authenticated publish.
Features
- HuggingFace-style discovery for agent resources, wrapped in a Win98 desktop with a real window manager (drag, minimize, z-order, taskbar, Start menu).
- Outcome filters, global search, leaderboard, Harness Heat, stars, server-side remix fork counters, threads, and verified gate-run counters.
- Harness detail opens as its own window with Overview, Install, Trust, Try sample, Thread, Files, and Versions tabs plus a plain-tone trust panel; Versions is backed by archive snapshot history.
- Authenticated quick publish flow (New Resource Wizard) that imports markdown into a small unverified scaffold.
- Share card window (harness_flex.exe), Wild West awards, Paint heat chart, and a paperclip mascot that opens the wizard.
- CLI package onlyharness with hh compatibility commands plus superskill install; HHREGISTRYURL targets another registry, default https://superskill.sh/api.
- Agent-friendly discovery: /llms.txt, /api/openapi.json, /server.json, and /mcp.
- Local bounty flow: create/claim/deliver/accept work-state over the existing gate_escrow rail; paid is set only after a matching escrow purchase captures against the delivered gate receipt.
- Semantic PR review and quality gate sidecar API.
- Docker production stack with system Caddy deployment mode for shared VPS hosts.
Architecture
flowchart LR
Web["React/Vite UI"] --> API["Fastify Harness API"]
API --> Seeds["Seed harnesses"]
API --> Imports["Imported harnesses"]
Web --> Supabase["Supabase Auth + RLS tables"]
API --> SupabaseAuth["Supabase token verification"]
Caddy["System Caddy / superskill.sh"] --> Web
CLI["hh CLI"] --> Schema["Schema, eval, gate, diff"]Run locally
npm install
npm run seed
npm run check
npm run smoke
npm run devOpen:
- UI: http://127.0.0.1:5177
- API: http://127.0.0.1:8787/healthz
- Local Gitea forge: http://127.0.0.1:3000
Operator Payouts
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
SuperSkill Registry: common questions
- Is SuperSkill Registry MCP server safe?
- Mostly: it is graded B (73/100). Read the SuperSkill Registry safety report
- How do I install SuperSkill Registry?
- It runs remotely at superskill.sh. Add it to Claude Code, Cursor, VS Code or Claude Desktop with the snippets above, or call it through the mcp.market gateway without installing anything.
- Does SuperSkill Registry need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is SuperSkill Registry maintained?
- The last commit was 96 days ago (2026-07-31). The latest release is v0.3.1.
- Is SuperSkill Registry up?
- 96% of our last 28 checks got an answer. We check remote servers about four times a day.
- What can I use instead of SuperSkill Registry?
- Servers from other publishers that do the same job: agent-device MCP server, AKShare MCP server and Claude Prompts MCP server.
- How does SuperSkill Registry compare with other AI and agents MCP servers?
- safety score 73/100 (higher than 42% of 4,460 AI and agents servers); tools 12 (the typical AI and agents server has 6); tools that change things 17% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to SuperSkill Registry
Same job from other publishers: the closest match first, then the best rated.
- agent-deviceLet AI agents inspect, control, and debug real iOS, Android, desktop, and TV appsnot reviewedWidely usedA
- AKShareLet a model explore AKShare's 1000+ China market data functions on its own — search, inspect, callnot reviewedEstablishedA
- Claude PromptsReusable prompt templates, multi-step workflow chains, and quality gates for any MCP client.not reviewedEstablishedA