Skip to main content
The sixtyfour-ai/skills repo packages Sixtyfour as an Agent Skill: one directory of instructions and a polling script that teaches an AI coding assistant to call the Sixtyfour API correctly. The skill calls the REST API with your API key. To give an assistant tools instead of instructions, use the MCP Server.

​
Prerequisites

  1. Have an AI coding assistant installed (Claude Code, Cursor, etc.).
  2. Sign up at app.sixtyfour.ai.
  3. Generate a key by following Get an API Key.
  4. Set SIXTYFOUR_API_KEY in your agent’s shell.

​
Install

You can install the Sixtyfour skills from Smithery, from the Cursor Plugin Directory, or manually from the sixtyfour-ai/skills repo.

​
Smithery

Open the Sixtyfour Intelligence skill on Smithery, pick your client, and run the install command Smithery generates.

​
Cursor Plugin Directory

Open the Cursor Plugin Directory listing and use the install button.

​
Manual install

The Sixtyfour Agent Skills are not yet available in the Claude Code marketplace. For now, follow the instructions below to install manually.
After cloning into Cursor, open the Command Palette (Cmd+Shift+P on macOS, Ctrl+Shift+P on Windows/Linux) and run Developer: Reload Window. Then open Cursor Settings → Plugins and confirm Sixtyfour appears with a Local badge and 1 skill.

​
Usage

Once installed, ask your assistant in plain English:

​
Troubleshooting

​
API key not set

Check that the variable is exported in the same shell your agent runs in:
If it’s not set, generate a key at app.sixtyfour.ai/keys and export it in the terminal that your agent runs in:
Setting the variable in a different terminal window does not carry over.

​
Environment variables don’t persist in Cursor or VS Code

Cursor and VS Code do not inherit shell environment variables when launched from the macOS Dock or Spotlight. Fix it one of three ways:
  • Launch from a terminal that already has the variable set: cursor .
  • Add the export to ~/.zshrc or ~/.bashrc so it’s always available
  • Add it to .claude/settings.json under env

​
Locating poll_job.py

The polling script ships inside the skill. When the skill is installed as a Claude Code plugin, it lives in the plugin cache:
For other installs, it is at scripts/poll_job.py inside the skill directory. It requires Python 3 and uses only the standard library.

​
429: too many concurrent searches

Deep search allows 5 concurrent searches per organization. If you hit a 429, wait for running searches to finish before starting new ones. Check running tasks with GET /search/status/{task_id}.

​
403 on high, xhigh, or scout tier requests

The high, xhigh, and scout tiers are enabled per organization. If you get a 403, book a call to request access. The micro, low, and medium tiers work for every organization.

​
Results are empty or incomplete

  • Bump tier from "low" to "medium". The default single-pass lookup misses people with limited online presence or common names.
  • For people intelligence, pass as many identifiers as possible in lead_info: full_name, company, linkedin_url, and email all improve matching.

​
Deep search appears stuck

Searches take 30–120 seconds or longer. The poll_job.py script prints progress messages like Analyzing your query... and Found X profiles as it goes. Don’t kill it early. If it hangs past 10 minutes, re-run with a fresh start-deep-search call.

​
Output file is CSV, not JSON

Deep search results download as CSV. Parse them with csv.DictReader, not json.load. Columns include fullName, currentPosition, currentCompany, linkedinUrl, skills, locationName, and others.

​
Signed download URLs expire

Download URLs expire 15 minutes after they are issued. To get a new URL, poll the status endpoint again and use the returned resource_handle_id.