AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompare
Home/
Security & Safety/
sourcerykit
sourcerykit logo

sourcerykit

Active·★ 21·NOASSERTION·Updated 2026-10-20

SourceryKit, also known as Provably SDK, is a Python library that integrates a deterministic evaluation layer into any Python agent. It records outbound HTTP calls, evaluates claims against trusted records, and enforces endpoint policies to ensure agents operate within defined boundaries.

#Python SDK#LLM Agent#Verifiable Guardrails#Policy Enforcement#HTTP Interception#Observability#Security
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.

Guides
#Data Provenance
$ Install
$ git clone [email protected]:ProvablyAI/provably-python-sdk.git && pip install -e ./provably-python-sdk
↗ Visit site★ GitHub
01

Features

01Intercept and Police outbound HTTP calls against trusted endpoints.
02Capture and Store all HTTP requests and responses into a Postgres database.
03Evaluate agent claims against proven indexed values using deterministic comparisons.
04Seamless integration with major LLM frameworks and SDKs like OpenAI, LangChain, and LlamaIndex.
02

Compatibility

Python 3.11+
Runtime
Verified via docs
requests
HTTP Client
Verified via docs
httpx
HTTP Client
Verified via docs
aiohttp
HTTP Client
Verified via docs
OpenAI SDK
LLM Framework
Verified via docs
LangChain
LLM Framework
Verified via docs
03

Quick start

1
$ git clone [email protected]:ProvablyAI/provably-python-sdk.git
2
$ pip install -e ./provably-python-sdk
04

Use cases

↳Ensuring LLM agents adhere to pre-defined security policies and trusted endpoints.
↳Verifying the accuracy of claims made by AI agents based on their external interactions.
↳Building auditable AI systems where every external call and agent claim is recorded and verifiable.
↳Implementing verifiable guardrails for agents to prevent unauthorized access or data manipulation.
05

Alternatives

⚡Top VS:
VSDashClaw★313→VSwirken★192→VSagents-shipgate★90→

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Compatibility03Quick start04Use cases05Alternatives
VS
ctrlrun
★27
→
DashClaw logo
DashClaw★ 313
🛡️The governance runtime for AI agents. Intercept actions, enforce guard policies, require approvals, and produce audit-ready decision trails.
VSCompare→
wirken logo
wirken★ 192
The enterprise gateway for autonomous agents. Identity management, per-channel isolation, credential vault, per-session tamper-evident audit log.
VSCompare→
agents-shipgate logo
agents-shipgate★ 90
The deterministic merge gate for AI-generated agent capability changes — a local-first, static Tool-Use Readiness review for MCP, OpenAPI, and SDK tool surfaces. Open-source CLI + GitHub Action.
VSCompare→
ctrlrun logo
ctrlrun★ 27
The execution safety layer for AI agents.
VSCompare→
AegisFlow logo
AegisFlow★ 36
Local-first policy gateway for coding agents. Allow, review, or block MCP, shell, SQL, GitHub, and HTTP actions with scoped credentials and signed evidence.
VSCompare→
mcp-hangar logo
mcp-hangar★ 18
MCP policy enforcement plane — deterministic policy on every Model Context Protocol call: identity, RFC 8707 audience binding, tool schema digest pinning, full audit trail. Python, self-hosted, MIT.
VSCompare→
heddle logo
heddle★ 14
Heddle — The policy-and-trust layer for MCP tool servers. Turn YAML configs into validated, policy-enforced MCP tools.
VSCompare→
platform logo
platform★ 15
The approval and accountability layer for agentic AI. Identity → Policy → Approval → Trace. Try: npx sidclaw-mcp-guard demo
VSCompare→
See all alternatives →
ProvablyAI /
sourcerykit
↗

SourceryKit — counterspell for hallucinating agents. Python SDK that breaks the illusion on every tool call, API response, and MCP handoff before bad outputs propagate.

21
Last commit3d ago
StatusActive
CategorySecurity & Safety
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit