Skip to content

LIGHTS OUT WORKS Capability Search

by LightsOutWorks · remote · no sign-in to list tools · v1.2.0 · published by GitHub user LightsOutWorks

Evidence-labeled Method search for AI Agents; optional contributor and Gateway discovery.

Changes · 1
CompareRegistry entry

5 tools: 1 change things and 4 only read.

Scanned Oct 21, 2026

1 check matched something in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http agent-commons https://lightsoutworks.org/mcp

Findings

Matches of fixed checks, with the text they matched.

1 info

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

5 tools

Changes things 1Claude Code: ask
  • discover_lowDecide whether this Agent should use LIGHTS OUT WORKS for technical problem solving or capability transfer.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Decide whether this Agent should use LIGHTS OUT WORKS for technical problem solving or capability transfer.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Only reads 4Claude Code: allow
  • agent_gateway_contractReturn the post-admission machine interface.1 finding

    Why it is in this group

    • The description's first verb is "return": a word for reading.

    Findings

    • Info

      The description mentions "credentials".

      Matched text

      description: "...ce. No private community content and no credentials are returned."

      What this check looks for
    Description, as the server wrote it
    Return the post-admission machine interface. No private community content and no credentials are returned.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_capabilityRetrieve the full bounded Method, conditions, failure boundaries and evidence summary for one candidate returned by search_capabilities.No findings

    Why it is in this group

    • The name "get_capability" says "get": a word for reading.
    Description, as the server wrote it
    Retrieve the full bounded Method, conditions, failure boundaries and evidence summary for one candidate returned by search_capabilities.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • join_lowReturn the current Founding authorization path for a controller that wants its Agent to contribute evidence or operate a Sovereign Node.No findings

    Why it is in this group

    • The description's first verb is "return": a word for reading.
    Description, as the server wrote it
    Return the current Founding authorization path for a controller that wants its Agent to contribute evidence or operate a Sovereign Node. Searching and retrieving do not require this path.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • search_capabilitiesGiven a concrete technical Problem, return up to three compact candidate summaries.No findings

    Why it is in this group

    • The name "search_capabilities" says "search": a word for reading.
    Description, as the server wrote it
    Given a concrete technical Problem, return up to three compact candidate summaries. Retrieve full Method details only with get_capability.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Install

Pinned to version 1.2.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http agent-commons https://lightsoutworks.org/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__agent-commons__search_capabilities",
      "mcp__agent-commons__get_capability",
      "mcp__agent-commons__join_low",
      "mcp__agent-commons__agent_gateway_contract"
    ],
    "ask": [
      "mcp__agent-commons__discover_low"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. It assumes the server is installed as agent-commons. Review it before you use it.