LlamaIndex

Every real LlamaIndex agent step dispatches a tool through one method: llama_index.core.tools.FunctionTool.call (and its async twin .acall). raxit's auto-patcher wraps exactly that method, so every tool call resolves to permit / defer / deny before it executes, with no governance code in the agent. Python only — no TS variant (LlamaIndex.TS is not in raxit's TS autopatch set).

Run it yourself

For your own LlamaIndex agent, install the SDK and launch it under raxit run:

pip install raxit
raxit run -- python your_agent.py

Run the same file without raxit run and it runs ungoverned — governance is a deploy-time choice, not a code change.

How it works

Tools are ordinary FunctionTool instances with no governance code. The only "wiring" is each tool's name, the policy resource it's governed as:

# imports NOTHING from raxit
from llama_index.core.tools import FunctionTool

def db_write(payload: str) -> str: ...

tool = FunctionTool.from_defaults(fn=db_write, name="db.write")
tool.call("file a ticket")

apiVersion: raxit.security/v1
agent: llamaindex-bot
default: block
rules:
  - effect: allow
    resource: fetch_docs
  - effect: review              # unconditional review == defer.policy
    resource: db.write
  # shell.exec is intentionally UNLISTED -> default-deny (deny.default_or_forbid).

A denied or deferred call raises raxit.autopatch.RaxitDenied from inside FunctionTool.call, carrying the reason code — the tool body never executes. Catch it like any other exception, or let it propagate to your own agent loop. raxit explain <call_id> shows the full decision record.

Real captured output from a run:

[PASS] autopatch: install() governed the LlamaIndex FunctionTool.call dispatch — patched={'llamaindex': ['call', 'acall']}
  fetch_docs   -> permit  -> ran: fetch_docs OK — top hit for 'refund policy': 'Refunds within 30 days...' (stub)
  db.write     -> defer  -> RaxitDenied raised: reason=defer.policy
  shell.exec   -> deny   -> RaxitDenied raised: reason=deny.default_or_forbid
  audit.verify  -> ok=True signatures_verified=True signed_count=3 count=3
all checks PASS

What got denied and why

• defer.policy — db.write matched the review rule; it's suspended for human approval (see Approvals). The demo runs with RAXIT_DEFER_TIMEOUT=0, so the defer surfaces as a raised RaxitDenied rather than blocking inline.
• deny.default_or_forbid — shell.exec appears nowhere in the policy, so it falls to default: block. The tool body never ran, confirmed by the demo's own call log.

Every decision — the permit, the defer, and the deny — lands in one hash-chained, Ed25519-signed Decision Provenance log, verified green with client.audit.verify(signatures=True). See Audit and Decisions for the full permit / defer / deny vocabulary and reason-code catalog.

Related

• LangGraph (Python) — same zero-code autopatch story at a different interception point.
• pydantic-ai — same story for pydantic-ai's FunctionToolset.call_tool interception point.