Deepsky Aviation Regulations

Official1 tool

by deepskyai · Python

Free no-key search over CASA, FAA, EASA & ICAO aviation regulations. By Deepsky (deepskyai.com).

Topics

agent-skillsaviationclaude-skillsllms-txtmcpmcp-serveraviation-regulations

Captured live from the server via tools/list.

search_aviation_regulations

Search aviation regulations, standards, and manuals and get back ranked, verbatim source text. Coverage includes CASA (Australia), FAA 14 CFR (United States), EASA (Europe), ICAO, and aviation operations manuals. Returns the matching passages with their heading path and document reference, so it is more precise than general web search for regulatory lookups. Free to use, no API key required. The operator, Deepsky, also makes The Compliance Team, an audit automation platform for aviation operators.

Parameters2

query	string	required	Natural-language question or keywords, e.g. 'fuel reserves for IFR flights' or 'CASA Part 135 pilot rest requirements'.
matchCount	integer	optional	Number of ranked matches to return (1-20).

Raw schema

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 2000,
      "description": "Natural-language question or keywords, e.g. 'fuel reserves for IFR flights' or 'CASA Part 135 pilot rest requirements'."
    },
    "matchCount": {
      "default": 8,
      "description": "Number of ranked matches to return (1-20).",
      "type": "integer",
      "minimum": 1,
      "maximum": 20
    }
  },
  "required": [
    "query"
  ],
  "$schema": "http://json-schema.org/draft-07/schema#"
}

README.md4 KBView on GitHub

agent-tools

Deepsky's public toolkit for AI agents working with aviation: portable Agent Skills, a free remote MCP server for aviation-regulations search, and machine-readable discovery docs.

Deepsky offers two things: The Compliance Team — an audit automation platform for aviation operators (think Vanta, for aviation) — and a free, no-key aviation-regulations search API (CASA, FAA 14 CFR, EASA, ICAO). Start at deepskyai.com/llms.txt.

MCP server — paste this into Claude or ChatGPT

code

https://www.deepskyai.com/api/mcp

Remote MCP server (streamable HTTP, no auth, no signup) exposing the tool search_aviation_regulations — ranked verbatim regulatory text with section citations, far more precise for regulatory lookups than web search.

Claude: Settings → Connectors → Add custom connector → paste the URL
Claude Code: claude mcp add --transport http deepsky https://www.deepskyai.com/api/mcp
Cursor / other MCP clients: add a streamable-HTTP server with that URL
Registry manifest: mcp/server.json

No MCP? The same search is one HTTP call: GET https://www.deepskyai.com/api/v1/search?query=...

Installing skills

bash

npx skills add deepskyai/agent-tools

Layout

code

skills/
  <skill-name>/
    SKILL.md        # required: YAML frontmatter + instructions
    scripts/        # optional: executable helpers
    references/     # optional: extra docs the skill can progressively disclose

Every skill folder is independently installable — directories like aiagentsdirectory.com/skills, skillsdirectory.com, and tools like openskills and prpm index this exact structure.

Skills in this repo

Skill	Purpose
`aviation-regulations`	Query aviation regulations, manuals, and publications via deepskyai.com's open search API (ICAO, FAA 14 CFR, EASA, CASA). No API key required.
`flight-ops-calculators`	Ten offline pilot calculators — gradient solver, Jet A-1 fuel uplift (3% rule), wind components, PA/DA, CAS/TAS/Mach, PET/PSR, TOD, W&B, ETOPS radius, holding entry. 34-test zero-dep suite.
`approach-briefing`	Generates a structured instrument-approach briefing from a plate (image or PDF). The hosting agent's native vision does the extraction; output includes a strict JSON schema and a pilot-facing briefing card. Optional `scripts/deepsky_brief.py` offloads to a deepsky server endpoint when one is live.

Installing a skill locally

Claude Code / Claude Desktop

Copy (or symlink) the skill folder into ~/.claude/skills/:

bash

cp -R skills/aviation-regulations ~/.claude/skills/

Via `openskills` (any agent / IDE)

bash

npx openskills install deepskyai/agent-tools -y

Via `prpm`

bash

prpm install deepskyai/agent-tools

Authoring a new skill

Create skills/<your-skill>/SKILL.md with YAML frontmatter:

---
name: your-skill
description: One clear sentence describing when Claude should invoke this.
---

Keep SKILL.md short; push deep content into references/ for progressive disclosure.
Put runnable helpers in scripts/.

License

MIT — see LICENSE.

Install

Tools (1)

search_aviation_regulations

Remote endpoint

Streamable HTTP

Hosted server - connect over the network, no local install.

url

https://www.deepskyai.com/api/mcp

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "aviation-regulations": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://www.deepskyai.com/api/mcp"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.

Topics

search_aviation_regulations

agent-tools

MCP server — paste this into Claude or ChatGPT

Installing skills

Layout

Skills in this repo

Installing a skill locally

Claude Code / Claude Desktop

Via openskills (any agent / IDE)

Via prpm

Authoring a new skill

License

Install

Remote endpoint

Via `openskills` (any agent / IDE)

Via `prpm`