mcpField
S
searching-packages

searching-packages

github.com/robomotionioDevelopment5121 installsUpdated 5d ago
Repository

About searching-packages

A Robomotion skill for finding packages, nodes, and templates and inspecting their schemas and documentation. It guides flow authors through the `robomotion` CLI, with fuzzy and synonym-expanded search across sources.

Key features

  1. 01Search packages, nodes, and templates
  2. 02Filter resources by package, tag, or category
  3. 03Inspect node schemas and package details
  4. 04Read and filter package documentation
  5. 05Fuzzy and synonym-expanded search

Use cases

  1. 01Find a node for a flow
  2. 02Check node properties before wiring
  3. 03Look up package authentication guidance

MCP support

The skill uses the `robomotion` CLI; search runs `robomotion-sdk-mcp` in the background. No separate MCP client is required.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/robomotionio/agent-skills --skill searching-packages

After installing, type/searching-packagesin Claude Code to confirm it loaded.

Bundled filesconfig.json

Details

From the project README

Searching Robomotion Packages

Discover Robomotion packages, nodes, and templates. All lookups go through the robomotion CLI — no separate MCP client required.

Four verbs cover every lookup

robomotion search <query>                        # cross-source fuzzy+semantic search
robomotion get <resource> [query] [flags]        # list or filter a resource
robomotion describe <resource> <name>            # full detail of one thing
robomotion docs <namespace>                      # read a package's llms.txt

Resources for get: packages | nodes | templates | categories | tags (plus vaults | vault-items | robots for runtime data). Resources for describe: node | package | template.

Typical workflow when authoring a flow

  1. Find what's available — if you don't know the package/node yet:

    robomotion search "http request"              # mixed results across sources
    robomotion get nodes click                    # nodes with "click" in the name
    robomotion get nodes --in Core.Browser        # every node in one package
    robomotion get templates bmi                  # match by slug/name/summary
    robomotion get templates --category "Web Scraping"
    robomotion get templates --tag api
    robomotion get categories                     # vocabulary
    robomotion get tags                           # vocabulary
    
  2. Verify properties before writing:

    robomotion describe node Core.Net.HttpRequest
    robomotion describe node Core.Net.HttpRequest,Core.Flow.GoTo   # batch
    

    Returns the full card: property names, types, descriptions, enums, defaults, and common properties (delayBefore, continueOnError, …).

  3. Read package docs when wiring a non-Core.* package:

    robomotion docs Robomotion.WordPress                   # compact (first 100 lines)
    robomotion docs Robomotion.WordPress --full            # whole llms.txt
    robomotion docs Robomotion.WordPress --grep auth       # line-filtered
    robomotion docs Robomotion.WordPress --section CreatePost
    

    docs carries auth patterns and package-level gotchas that aren't in node schemas. Run it for every non-Core.* package you use. Many packages have no llms.txt (for example Robomotion.HermesAgent): then docs answers "available":false, and the package's own descriptions are the documentation — read robomotion describe package <ns> for its nodes and versions, and robomotion describe node <type> for each node you wire (every property carries a description).

Search engine capabilities

Under the hood robomotion search uses Bleve full-text search with:

FeatureBehavior
Full-textBM25 ranking across title, name, description, content, keywords
FuzzyLevenshtein ≤ 2 — tolerates typos (browsr → browser, wordpres → wordpress)
SemanticSynonym expansion — browser also matches chrome, firefox, web, scraping
Multi-sourcePackage docs (llms.txt) and templates in one call

Built-in synonym families:

  • browser ↔ chrome · firefox · web · scraping
  • automation ↔ robot · flow · workflow · process
  • wordpress ↔ blog · cms · post · article
  • database ↔ sql · sqlite · query · table
  • email ↔ mail · smtp · send · message
  • file ↔ filesystem · directory · folder · path

Filter flags

robomotion search "automation" --limit 5
robomotion get nodes "click" --package Core.Browser
robomotion get templates --tag api --limit 10
robomotion describe node Core.Browser.ClickElement --json     # raw MCP response

First-call warm-up

robomotion search spawns robomotion-sdk-mcp under the hood; its Bleve index takes ~3–6 s to build on cold start. The CLI retries automatically on "Search index not ready", so you'll see one slow search per session, then near-instant afterward.

Quick reference

I want to…Command
Find anything by keywordrobomotion search <query>
List all packagesrobomotion get packages
Find a packagerobomotion get packages <keyword>
List nodes in a packagerobomotion get nodes --in <namespace>
Find nodes by keywordrobomotion get nodes <keyword>
Get full node schemarobomotion describe node <type>
Read package docsrobomotion docs <namespace>
Grep package docsrobomotion docs <namespace> --grep <pattern>
List available vaultsrobomotion get vaults
List robotsrobomotion get robots

Related Skills

  • creating-flow — generate a flow from the nodes you found
  • validating-flow — pspec-check the result
  • running-flow — submit + stream agent events

Comments

Markdown supported