The decision-continuity protocol for human-to-agent work transfer.

// The cold-start problem

Most agentic AI failures aren’t reasoning failures. They’re cold-start failures.

The agent begins each session with no idea what the human already concluded, what tabs they had open, or which option they were leaning toward. UNIHODL exposes the human’s working session as a first-class, machine-readable artifact.

// The four primitives

Clean, orthogonal, each with one job.

// Quickstart · 5 minutes

Three steps from install to a Claude agent that picks up mid-thought.

npm install @unihodl/agent-sdk
# or
pip install unihodl-agent

curl https://unihodl.app/api/v1/resume_tokens \
  -H "Authorization: Bearer $UNIHODL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "session_id": "ses_8f3aZ91b",
    "scopes": ["read:context", "read:reasoning"],
    "ttl_seconds": 3600,
    "audience": "claude.anthropic.com"
  }'

from unihodl_agent import Client
from anthropic import Anthropic

uh = Client(api_key=os.environ["UNIHODL_API_KEY"])
claude = Anthropic()

# Hydrate the session — returns a structured Resume Context
ctx = uh.sessions.hydrate("ses_8f3aZ91b")

resp = claude.messages.create(
    model="claude-sonnet-4-7",
    max_tokens=2048,
    system=ctx.as_system_prompt(),
    messages=[
        {"role": "user", "content": "Continue Sarah's research on API v3."}
    ],
)
print(resp.content[0].text)

// What you get

Six capabilities. Cross-framework guarantee.

// Reference integrations

Drop-in for the four agent frameworks people actually use.

// Design principles

Audience-bound. Redacted at the boundary. Auditable.

• Audience-bound by default.Every resume token is bound to a single audience (a model vendor, an MCP server URL, or a known agent identity). Tokens cannot be replayed against other audiences.
• Redaction at the boundary.Sensitive content is redacted server-side before serialization based on per-workspace policies — the agent never sees content it isn't entitled to.
• Reasoning is structured, not narrative.Decision threads are serialized as typed graph nodes, not free text, so agents can ground tool calls in specific intent vectors.
• Format-pluggable.The same Resume Context can serialize to a Claude system prompt, a Gemini system instruction, an OpenAI Agents SDK input, or a raw MCP resource.
• Auditable.Every hydration creates an immutable access record: who, when, what scopes, what was redacted, what was returned.