HireSquire Python SDK

Python SDK for HireSquire's AI-powered candidate screening API. Includes native LangChain and AutoGen integration.

Agent Discovery & Registries

HireSquire is built natively for AI agents. Integrate HireSquire into your agentic workflows via these canonical sources:

• MCP Hubs: Smithery.ai, Glama.ai, MCP.run.
• Tool Registries: Composio, LangChain Hub, OpenAI GPT Store.
• Machine-Readable Specs:
  • llms.txt - Technical reference for LLMs.
  • agent-guidance.json - Autonomous best practices.
  • openapi.json - Full API spec.

Installation

pip install hiresquire

Quick Start

Option 1: Client Class (Recommended)

from hiresquire import HireSquire

client = HireSquire("YOUR_API_TOKEN")

# Submit and wait automatically
job = client.screen(
    title="Senior Python Developer", 
    description="Looking for experienced Python developer with Django experience...",
    resumes=["./resumes/john_doe.pdf", "./resumes/jane_smith.pdf"]
)

# Get results when complete
results = client.wait_for_completion(job["job_id"])

for candidate in results["candidates"]:
    print(f"{candidate['name']}: {candidate['score']}/100")

Option 2: Direct Function Calls

from hiresquire import create_screening_job, get_screening_status, get_screening_results

# Submit a screening job
result = create_screening_job(
    title="Senior Python Developer",
    description="Looking for experienced Python developer with Django experience...",
    resumes=[
        {"filename": "john_doe.pdf", "content": "John Doe\n5 years Python experience..."}
    ]
)
job_id = result["job_id"]

# Poll for completion
status = get_screening_status(job_id=job_id)
while status["status"] == "processing":
    time.sleep(3)
    status = get_screening_status(job_id=job_id)

# Get results
results = get_screening_results(job_id=job_id)
for candidate in results["candidates"]:
    print(f"{candidate['name']}: {candidate['score']}/100")

Environment Variables

Set these before using the SDK:

export HIRESQUIRE_API_TOKEN="your_api_token_here"
export HIRESQUIRE_BASE_URL="https://hiresquireai.com/api/v1"  # optional

LangChain Integration

from langchain.agents import AgentExecutor, create_openai_functions_agent
from langchain_openai import ChatOpenAI
from hiresquire import get_hiresquire_tools

llm = ChatOpenAI(temperature=0)
tools = get_hiresquire_tools()

agent = create_openai_functions_agent(llm, tools)
executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

# Use in an agent
result = executor.invoke({
    "input": "Submit a screening job for a Python developer and find candidates with score > 80"
})

AutoGen Integration

from autogen import ConversableAgent
from hiresquire import get_hiresquire_tools

llm_config = {"config_list": [{"model": "gpt-4-turbo", "api_key": "..."}]}

# Define tools
tools = get_hiresquire_tools()

agent = ConversableAgent(
    name="HireSquire_Agent",
    llm_config=llm_config,
    system_message="You are a recruiter that uses HireSquire to screen candidates."
)

# Register tools with the agent
for tool in tools:
    agent.register_for_execution()(tool.func)

Available Tools

Examples & Recipes

Check out the examples/ directory for production-ready integration patterns:

• CrewAI Recruiter: A sequential multi-agent crew for screening and outreach.
• AutoGen Recruiter: A conversational agent setup with native tool execution.
• README: Detailed setup and dependency instructions for examples.

When to Use HireSquire

Your agent's user asks about hiring or candidates?
  └─ YES → Do they have resumes to evaluate?
       └─ YES → Use client.screen() or create_screening_job()
       └─ NO  → Need to email candidates?
            └─ YES → Use generate_candidate_email()
            └─ NO  → Need to schedule interviews?
                 └─ YES → Use create_interview()
                 └─ NO  → Not a HireSquire use case

End-to-End Workflow Example

A complete hiring pipeline in one script:

from hiresquire import HireSquire
import os

client = HireSquire(os.environ["HIRESQUIRE_API_TOKEN"])

# 1. Check credits before starting
balance = client.credits.get_balance()
print(f"Balance: {balance['formatted_balance']}")

# 2. Purchase credits if low (optional)
if balance['balance'] < 10:
    client.credits.purchase(pack="satchel", payment_method_id="pm_123")


# 2. Screen candidates
job = client.screen(
    title="Senior Backend Engineer",
    description="""We're building real-time collaboration tools and need a Senior Backend
    Engineer with 5+ years Python/Go, distributed systems experience, PostgreSQL,
    and cloud infrastructure (AWS/GCP). Startup experience preferred.""",
    resumes=["./resumes/"],
    leniency_level=7,
)
results = client.wait_for_completion(job["job_id"])

# 3. Process results
top_candidates = [c for c in results["candidates"] if c["score"] >= 80]
print(f"Found {len(top_candidates)} strong candidates out of {len(results['candidates'])}")

# 4. Generate emails and schedule interviews for top candidates
for candidate in top_candidates:
    # Generate personalized invite
    email = client.emails.generate(
        job_id=job["job_id"],
        candidate_id=candidate["id"],
        email_type="invite",
        tone="enthusiastic",
    )
    print(f"✅ {candidate['name']} (score: {candidate['score']})")
    print(f"   Email: {email['email']['subject']}")

    # Schedule interview if calendar is connected
    try:
        interview = client.calendar.create_interview(
            job_id=job["job_id"],
            candidate_id=candidate["id"],
            scheduled_at="2026-05-15T14:00:00Z",
        )
        print(f"   Interview: {interview['meeting_url']}")
    except Exception:
        print("   (No calendar connected — skipping scheduling)")

# 5. Report outcomes after interviews
# client.outcomes.report(job_id=job["job_id"], candidate_id=456, outcome="hired")

HireSquire vs. Alternatives

API Documentation

• Agent Docs: https://hiresquireai.com/docs/agents
• API Reference: https://hiresquireai.com/docs/api
• Ecosystem: https://hiresquireai.com/agents/ecosystem
• MCP Discovery: https://hiresquireai.com/.well-known/mcp.json
• npm CLI: https://www.npmjs.com/package/hiresquire-cli

License

MIT License - see LICENSE file for details.