# Agent Harness Hackathon 2026 — What I built, how I used TrueForge, and what I learned

> Source: <https://dev.to/parth_pinjarkar_86f7f3ec2/agent-harness-hackathon-2026-what-i-built-how-i-used-trueforge-and-what-i-learned-5fh6>
> Published: 2026-08-30 09:08:14+00:00

THE PROBLEM

Ransomware attacks cost organizations over $20 billion every year. Security teams are drowning in 11,000+ alerts per day, most of which are false positives. By the time a human analyst investigates, the encryption has already spread to hundreds of files.

The average response time? Hours. The average recovery cost? $2.73 million.

I wanted to build something that could change that — an AI agent that doesn't just detect threats, but investigates and responds to them autonomously, while still keeping a human in the loop for irreversible actions.

That's how AutoVault was born.

WHAT I BUILT

AutoVault is an autonomous AI security operations agent that runs on TrueForge, the open-source agent harness by TrueFoundry. It detects ransomware in 0.3 seconds, spawns 5 parallel subagents to investigate, and responds autonomously — but always stops to ask a human before doing anything irreversible.

HOW I USED TRUEFORGE

TrueForge isn't just a dependency — it's the brain of the entire system. I used every single TrueForge feature, all 12 of them.

MCP TOOLS

I built 4 Python MCP servers using the MCP SDK v2:

Each tool has full input schemas, proper error handling, and returns structured JSON.

SUBAGENTS

TrueForge's subagent system was a game-changer. I spawned 5 specialized agents that run in parallel:

The speed improvement was dramatic — what took hours now takes seconds.

SANDBOX (DAYTONA)

I wrote 20 Python scripts that run safely in TrueForge's Daytona sandbox:

HUMAN APPROVALS

TrueForge's approval system lets me define 10 rules with risk scoring:

The 7-step workflow ensures nothing irreversible happens without a human saying yes.

SKILLS

I created 4 instruction packs that TrueForge loads dynamically:

CONTEXT ENGINEERING

TrueForge's context engineering kept the agent efficient:

PERSISTENT SESSIONS

The agent saves its state to SQLite every 30 seconds. If the connection drops mid-investigation, it resumes exactly where it left off.

THE 13 INNOVATIONS

Human-Like Memory Architecture — Episodic, semantic, procedural, and working memory. The agent remembers past attacks.

Predictive Ransomware Defense — Predicts attacks 30 minutes before they happen.

Recursive Self-Improvement — The agent improves its own code and algorithms.

Collective Intelligence Swarm — Hive mind of agents making decisions together.

Neuromorphic Processing — Brain-inspired neural networks with sub-millisecond latency.

Self-Healing File System — Automatically restores encrypted files from blockchain backups.

AI Red Team vs Blue Team — Two AI teams compete and improve through adversarial training.

Knowledge Graph — Graph database mapping relationships between threats, actors, tools, techniques.

Federated Learning — Multiple organizations train models without sharing sensitive data.

Digital Twin Simulation — Virtual replica for risk-free security testing.

Adversarial Robustness — Defends against attacks designed to trick the AI.

Natural Language Threat Intelligence — Query threats using plain English.

Self-Learning Security Evolution — Detection models evolve using genetic algorithms.

THE QODO CODE REVIEW EXPERIENCE

Every change goes through Qodo review via GitHub Actions. On PR #1, Qodo found:

All three were real issues. I fixed all HIGH findings and documented everything in the README.

WHAT I LEARNED

TrueForge Is More Than a Wrapper — The MCP tool system, subagent orchestration, approval workflows, and context engineering are all deep, production-grade features.

Parallel Subagents Change Everything — Running 5 investigations simultaneously isn't just faster — it's qualitatively different.

Human Approval Is Non-Negotiable — For a security tool, having human approval is critical. TrueForge made this trivial to implement.

Qodo Actually Finds Bugs — It understands context and finds issues that surface-level analysis misses.

The Sandbox Is Your Safety Net — When running scripts that analyze potentially malicious files, you need a sandbox.

Documentation Matters — Writing the README forced me to think about what I actually built and why.

TRY IT YOURSELF

github.com/ItsParthPinjarkar/autovault

git clone [https://github.com/ItsParthPinjarkar/autovault.git](https://github.com/ItsParthPinjarkar/autovault.git)

cd autovault

pip install -r agent/requirements.txt

pip install -r mcp-server/requirements.txt

bash setup-trueforge.sh

python demo_advanced.py

ACKNOWLEDGMENTS

Built for the Agent Harness Hackathon (August 24-30, 2026).

This project was built with AI assistance. All code has been reviewed, tested, and understood by the developer.
