Use the marketplace

This guide is for people who want to use skills: find one, connect your coding agent, run it and manage your access.

#Find a skill

#On the website

Press Ctrl+K (⌘+K on macOS) or / anywhere on the site to open quick search.

#From the terminal

skillgild search "accessibility"

accessibility-audit      free     Accessibility audit — Reviews a UI description or component code…
design-review            USD 15.00 Design review — Structured critique of a screen or flow…

#From your agent

Once the MCP server is connected, your agent can call skillgild_search_skills itself. The tool returns public metadata, access tier, version and input schema; never the private implementation.

Search matches a case-insensitive substring of a skill's name, slug or description, and returns up to 100 results.

#Connect your agent

You set this up once per machine. After that, every skill is available to every connected agent.

#1. Install the CLI

The skillgild CLI is both the sign-in tool and the local MCP server your agent talks to. Install it with Go 1.26 or later from the SkillGild source:

cd skillgild/backend
go install ./cmd/skillgild

#2. Sign in from the terminal

skillgild login

Open https://skillgild.dev/connect/device?user_code=ABCD-EFGH
Enter code: ABCD-EFGH
Connected. SkillGild saved your revocable credential in the operating system credential store.

The CLI opens the device page in your browser. Sign in, confirm that the code matches, and approve it. The code expires after 10 minutes. The resulting API key is stored in your operating system's credential store under the service name SkillGild, never in a config file. The machine then appears under Account → Agent connections.

Run skillgild status at any time to confirm which account you are connected as.

#3. Register the MCP server with your agent

Find the absolute path of the binary with command -v skillgild, then register it. Apps launched from an IDE may not inherit your terminal's PATH, so use the absolute path.

{
  "mcpServers": {
    "skillgild": {
      "type": "stdio",
      "command": "/absolute/path/to/skillgild",
      "args": ["mcp"]
    }
  }
}

After setup, check that the skillgild server shows as connected in your agent's MCP settings.

#4. Install a skill

skillgild install accessibility-audit --agent claude-code

Installed hosted wrapper at /Users/you/.claude/skills/accessibility-audit/SKILL.md
The wrapper contains public metadata only. Connect the SkillGild MCP server to run the protected capability.

"Installing" writes a small SKILL.md wrapper, not the skill itself. It tells your agent when to use the skill and how to call it:

---
name: accessibility-audit
description: "Reviews a UI description or component code and returns prioritized WCAG 2.2 issues with fixes."
---

# Accessibility audit

Free hosted skill. SkillGild keeps the private implementation on its hosted service. This local
file is only a reference and invocation guide.

When the user's coding task matches this skill, use the SkillGild MCP tool `skillgild_run_skill`
with skill ID `<skill uuid>` and an input object conforming to this public input schema:
…

#5. Use it

Ask your agent for the task in plain language. When the task matches, the agent calls skillgild_run_skill with the skill ID and an input object, then applies the returned result with its own file tools. The wrapper instructs it to show a diff and ask before destructive changes.

You can also run a skill directly:

skillgild run accessibility-audit --input '{"prompt":"Audit this signup form: ..."}'

#Configure a skill's input

Every skill declares a public input schema. You can see it on the skill page, in the installed wrapper, and from GET /v1/skills/{slug}. Send a JSON object that matches it:

• include every key listed in required;
• use the declared top-level type for each property;
• many skills use the open default schema, where {"prompt": "..."} is the convention.

Input is limited to 64 KiB by default.

#Access, pricing and quotas

Every run response includes your current usage, so agents and scripts can tell how much allowance is left:

"usage": { "used": 3, "limit": 25, "reset_at": "2026-11-01T00:00:00Z", "access_tier": "free" }

Platform-wide safety limits also apply:

#Manage your access

Everything lives in Account:

• Agent connections → devices lists each connected device with its last use and credential expiry. Revoking a device disables its key immediately. skillgild logout does the same from the terminal and removes the local credential.
• Agent connections → API keys are for scripts and SDKs. You can have 5 active keys in total, including CLI device keys. A key's secret is shown once; keys expire after 90 days by default.
• Organizations & webhooks → webhooks send signed execution.completed and execution.failed events to your HTTPS endpoint. See Webhooks.
• Organizations & webhooks → organizations let a team share paid-skill access. Create a workspace, invite members with one-time tokens, and transfer ownership.

#Report a problem

Signed-in users can report a skill from its page with a reason of 40 to 4,000 characters. Reports are reviewed by the SkillGild team, and you can follow them under Account → Activity & support. You can appeal a closed decision within 30 days. For anything else, open a support case from the same section.

#Privacy at a glance

• Run inputs and outputs are not stored in execution records.
• Abuse review stores only HMAC fingerprints of inputs and coarse signals, never the request text.
• If you send an Idempotency-Key, the response is kept encrypted for 24 hours so a retry can replay it.
• Catalog search analytics count searches in aggregate, with no query text, account or IP address.