TapPass Documentation
Connect any agent to any model — with one governance layer you control.
Follow the path
Section titled “Follow the path”Four steps from an ungoverned call to full compliance evidence.
1. Onboard Install the SDK, point at your TapPass server, make one governed call. Five minutes.
2. Connect Plug in your providers, frameworks, IDEs, SSO, and data sources. Mostly zero code.
3. Govern Pipeline, flags, policy — decide what every call must pass before it reaches a model.
4. Prove Export GDPR Art. 30, EU AI Act Art. 14, and SOC 2 evidence straight from the audit trail.
Or jump to what you need
Section titled “Or jump to what you need” Integrations catalogue Providers, frameworks, IDEs, SSO, data, SIEM — one page lists them all.
Governance Pipeline The 49 steps that run on every call. Input validation → detection → policy → audit.
Python SDK The Agent class: chat, stream, async, govern, secure, reset.
HTTP API OpenAI-compatible REST for any language.
Quick install
Section titled “Quick install”pip install tappassfrom tappass import Agent
agent = Agent("https://tappass.example.com", "tp_...")response = agent.chat("What are the GDPR requirements?")print(response.content)Or, no code at all:
export OPENAI_BASE_URL=https://tappass.example.com/v1export OPENAI_API_KEY=tp_...Every call — from Cursor, Copilot, LangChain, CrewAI, or curl — now flows through your governance pipeline.