Open Source · MIT License · Zero Infrastructure

The Governance Layer
for AI Agents

Your AI agents work autonomously — but do you actually know what they're doing? Satya gives agents a task board, permission scope, and an audit trail humans can watch in real time.

Star on GitHub Quick Start →
Help us reach 200 stars — every star funds open-source AI tooling
MIT License
·
Python 3.11+
·
3 lines to integrate
·
Zero database / cloud required
·
Claude Code compatible
·
Built by AnkTechSol
How It Works

Agent operates. Human observes.

Satya enforces a clean separation: the AI agent owns all operations. The human gets a live browser dashboard — no terminal needed.

AI Agent (Operator)

  • Deploys Satya and starts the dashboard
  • Creates tasks with allowed & forbidden actions
  • Logs progress via the Python SDK
  • Checks permissions before every action
  • Can't self-certify completion — checker decides
  • Scrapes knowledge base from URLs

Human (Observer)

  • Opens the dashboard URL in a browser
  • Sees the real-time task board (To Do / In Progress / Done)
  • Reads timestamped agent session logs
  • Reviews scraped knowledge base
  • Monitors completion metrics & charts
  • Optionally creates tasks via the UI
The Problem

Every monitoring tool asks the wrong question.

LangSmith, Langfuse, AgentOps — they all tell you what your agent did. Nobody was asking what your agent is allowed to do.

Without Satya

Flying blind

Your agent has no task scope. It deletes files it wasn't supposed to touch, marks itself done, and charges you $40 in API costs.

  • No permission boundaries per task
  • Agent self-certifies completion
  • No real-time visibility for humans
  • No audit trail by default
  • Post-hoc logging only — damage already done
With Satya

Full accountability

You define the scope. Satya enforces it. Every action is logged before it happens. Humans monitor in real time with zero setup.

  • Per-task allowed_actions & forbidden_actions
  • Completion checker — not self-certified
  • Real-time dashboard for human oversight
  • Git-based audit trail (optional)
  • Watchdog detects stale tasks automatically
Quick Start

3 lines to full agent governance.

Drop into any Python agent in seconds. No database, no cloud, no config file needed.

Python SDK
Docker (1 command)
# 1. Install
pip install satya-ai

# 2. Use in your agent — 3 lines
import satya.sdk as satya

client = satya.init("my_agent")                              # start session
task   = client.create_task("Analyze pricing", "Compare competitors")  # create task
satya.log("Starting analysis...")                             # log progress

# Agent does the work...
client.update_task(task["id"], "Done")                     # mark done (checker validates)
client.flush_logs()                                         # persist via git (optional)

# Human opens http://localhost:5000 — sees everything above in real time.
# No terminal. No config. Just a browser.
Features

Everything you need. Nothing you don't.

Zero infrastructure. Flat JSON files. Git-compatible. Works with Claude Code, Jules, Codex — any agent with Python access.

📋

Real-Time Task Board

Kanban-style columns (To Do / In Progress / Done) with priorities. Agents update it. Humans watch it. No refresh needed.

🛡️

Action Validator

Per-task allowed_actions and forbidden_actions. The can_do() function blocks everything not explicitly allowed. Default-deny.

Completion Checker

Agents can't self-certify. Verify by file existence, test pass, subtask completion, or manual human approval.

🔍

Knowledge Base (Truth Source)

Scrape any URL to clean Markdown, stored in a shared knowledge base. Agents and humans both reference the same ground truth.

📜

Agent Session Logs

Timestamped entries from every connected agent, visible in real time on the dashboard. Full audit trail, zero setup.

👁️

Watchdog

Automatically detects stale tasks — tasks that haven't moved in too long. Surfaces them on the dashboard with a warning badge.

🔄

Agent Runner (Polling Loop)

Drop-in polling runner for Claude Code and compatible agents. Checks for new tasks, executes, reports — fully autonomous.

🗂️

Zero Config Storage

Everything lives in flat JSON + Markdown files in satya_data/. No database, no cloud service, no migration scripts.

🌙

Dark / Light Theme

One-click toggle in the sidebar. Full CSS variable swapping. Works great on the projector in your next demo.

Give Satya a ⭐ — it's free

Each star signals to the AI community that open-source agent governance matters. Help us hit 200 stars this week.

Star on GitHub — github.com/AnkTechsol/Satya_AI
MIT License Python 3.11+ Zero infrastructure Claude Code compatible