DEV Community

Lars
Lars

Posted on • Originally published at moltrust.ch

Add trust verification to your CrewAI and LangChain agents in one line

#crewai #langchain #python #aiagents

Before your agent calls another agent, it should probably check whether that agent is actually trustworthy. Right now, it doesn't.

This came up in crewAI/issues/4877 — a proposal for a GuardrailProvider interface that sits between the hook system and authorization logic. The gap is real: existing guardrails validate output after task completion. Tool-call authorization needs to happen before execution, per call, across all tasks.

We shipped a provider for it today.

CrewAI 

pip install moltrust-crewai
Enter fullscreen mode Exit fullscreen mode 
from moltrust_crewai import MolTrustGuardrail

guard = MolTrustGuardrail(min_score=60)
guard.install()  # registers before_tool_call hook
Enter fullscreen mode Exit fullscreen mode

Before every tool call, MolTrustGuardrail resolves the calling agent's DID, checks its behavioral trust score (0–100) against the MolTrust registry, and returns False to block if the score falls below your threshold. No API key required for read-only checks.

LangChain 1.x 

pip install moltrust-langchain
Enter fullscreen mode Exit fullscreen mode 
from moltrust_langchain import MolTrustMiddleware
from langchain.agents import create_agent

agent = create_agent(
    model="anthropic:claude-sonnet-4-6",
    tools=[...],
    middleware=[MolTrustMiddleware(min_score=60)],
)
Enter fullscreen mode Exit fullscreen mode

Hooks into before_model and wrap_tool_call. Blocking raises TrustCheckFailed; warning logs and continues.

Three modes 

MolTrustGuardrail(
    min_score=60,
    action="block",   # "block" | "warn" | "log"
)
Enter fullscreen mode Exit fullscreen mode

Start with warn during rollout, move to block when you understand your score distribution.

No account required to start

Tier 1 is keyless — the trust score endpoint is public and rate-limited.

curl https://api.moltrust.ch/skill/trust-score/{did}
Enter fullscreen mode Exit fullscreen mode

Tier 2 (free account) adds higher rate limits and lets your orchestrator report interaction outcomes back — that is where the feedback loop starts and scores get more precise over time.

How the score is computed

0–100, based on endorsement graph, behavioral history, Sybil detection, and on-chain signals, with time-decayed weighting. The score is recomputable — formula and on-chain inputs are public:

curl https://api.moltrust.ch/credits/solvency/{did}
# Formula: github.com/MoltyCel/moltrust-api/blob/main/docs/solvency-usdc-v0.md
Enter fullscreen mode Exit fullscreen mode

One note on Microsoft AGT

AGT does policy enforcement — what an agent is permitted to do. MolTrust does behavioral trust — is this agent trustworthy, based on verifiable history. Different questions, clean composition.

Source: github.com/MoltyCel/moltrust-crewai · github.com/MoltyCel/moltrust-langchain

Full write-up: moltrust.ch/blog/crewai-trust-middleware.html

Top comments (0)