惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
S
Securelist
Simon Willison's Weblog
Simon Willison's Weblog
Security Latest
Security Latest
博客园 - 司徒正美
T
Threat Research - Cisco Blogs
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
C
Check Point Blog
Jina AI
Jina AI
H
Heimdal Security Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Exploit-DB.com RSS Feed
Exploit-DB.com RSS Feed
B
Blog
美团技术团队
W
WeLiveSecurity
G
Google Developers Blog
K
Kaspersky official blog
Application and Cybersecurity Blog
Application and Cybersecurity Blog
C
Cybersecurity and Infrastructure Security Agency CISA
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
V
Visual Studio Blog
T
Threatpost
酷 壳 – CoolShell
酷 壳 – CoolShell
www.infosecurity-magazine.com
www.infosecurity-magazine.com
Attack and Defense Labs
Attack and Defense Labs
腾讯CDC
Hacker News - Newest:
Hacker News - Newest: "LLM"
Cyberwarzone
Cyberwarzone
Apple Machine Learning Research
Apple Machine Learning Research
P
Privacy & Cybersecurity Law Blog
H
Hacker News: Front Page
博客园 - 【当耐特】
人人都是产品经理
人人都是产品经理
有赞技术团队
有赞技术团队
Hugging Face - Blog
Hugging Face - Blog
Recent Commits to openclaw:main
Recent Commits to openclaw:main
D
Darknet – Hacking Tools, Hacker News & Cyber Security
C
CERT Recently Published Vulnerability Notes
V
Vulnerabilities – Threatpost
AI
AI
MongoDB | Blog
MongoDB | Blog
Project Zero
Project Zero
Hacker News: Ask HN
Hacker News: Ask HN
量子位
A
Arctic Wolf
GbyAI
GbyAI
Recorded Future
Recorded Future
SecWiki News
SecWiki News
小众软件
小众软件
Martin Fowler
Martin Fowler

Hacker News: Show HN

PurrrrrFocus: Pomodoro Timer App - App Store Workflow Engine — Multi-Step Orchestration for Bun RapidPhoto: Pro Photo Editor App - App Store GitHub - DheerG/swarms: Achieve extraordinary results with claude code across a variety of tasks SPICE simulation → oscilloscope → verification with Claude Code — Lucas Gerads Show HN: VCoding – A 5 MB native Windows IDE with no dynamic dependencies Show HN: LLMs don't hallucinate because they're bad at math, it's the format GitHub - Agent-FM/agentfm-core: AgentFM is a peer-to-peer network that turns everyday computers into a decentralized AI supercomputer. AgentFM lets you run massive AI workloads directly across a global mesh of idle CPUs and GPUs. Show HN: Tracking Top US Science Olympiad Alumni over Last 25 Years GitHub - Potarix/agent-hub: One place to talk to all your agents Show HN: Runtime security for AI agents(injection,tool abuse, data exfiltration) GitHub - dubeyKartikay/lazyspotify: Terminal Spotify client for macOS and Linux GitHub - the-banana-tool/king-louie: Easy to use GUI Personal AI Assistant. Win/Linux/Mac. Show HN I made my vacation rental bookable by AI agents–no Airbnb, 0% commission GitHub - basteez/jsf-autoreload: maven plugin to enable hot reload on jsf projects uvm32/hosts/host-gdbstub at main · ringtailsoftware/uvm32 GitHub - labsai/EDDI: Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, GDPR, HIPAA). Built on Quarkus. GitHub - glitchnsec/fortyone-oss: AI Executive Assistant Platform Quickstart | Alien GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. GitHub - ocrbase-hq/ocrbase: 📄 PDF/IMG ->.MD/JSON Document OCR API for PaddleOCR and GLMOCR. Self-hostable. GitHub - impactjo/home-memory: MCP server that lets your AI assistant remember everything about your home. GitHub - Sets88/dbcls: DbCls is a powerful terminal database client that supports various databases GitHub - neptun2000/heor-agent-mcp GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh RollQuation: Math Puzzles - Apps on Google Play GitHub - dropbox/witchcraft Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis GitHub - opentalon/opentalon: OpenTalon is an open-source platform built from the ground up in Go as a robust alternative to OpenClaw LinkedIn™ 职位抓取工具 - Chrome 应用商店 GitHub - EdoardoBambini/Agent-Armor-Iaga: AI agents are getting tool access — shell, file system, databases, APIs, secrets. But **nobody is governing what they actually do with it**. Frameworks like LangChain, CrewAI, AutoGen, and Claude Code give agents the power to execute. Agent Armor gives you the power to control, audit, and approve every single action before it happens. HN Vibes — Week 15, Apr 7–13 2026 GitHub - chojs23/ec: Easy terminal-native 3-way git mergetool vim-like workflow GitHub - SethPyle376/hiraeth: Local AWS emulator focused on fast integration testing, with SQS support, SQLite-backed state, and a debug-friendly web UI. GitHub - JakOb-dotcom/cloud-sandbox-security-analysis: Technical analysis and Proof of Concept (PoC) regarding environment variable exfiltration in containerized cloud sandboxes via side-channel data leaks. Springboards - Flint Alpha Show HN: A simpler coding agent harness GitHub - audiodude/sudomake-friends GitHub - 256thFission/mini-mythos: OSS clone of Anthropic’s Mythos harness to locate C/C++ memory vulnerabilities Show HN: OpenParallax: OS-level privilege separation for AI agent execution Hacker News Sorted - Chrome 应用商店 Show HN: How to Install Docker on Ubuntu 24.04 LTS: Complete 2026 Guide GitHub - himanshudongre/smriti GitHub - sverrirsig/claude-control: macOS desktop dashboard for monitoring and managing multiple Claude Code sessions GitHub - ory/dockertest: Write better integration tests! Dockertest helps you boot up ephermal docker images for your Go tests with minimal work. Chiral - Chrome 应用商店 Show HN: Two Claudes collaborating through shared memory on a $100 mini-PC GitHub - pmichaillat/latex-cv: Minimalist LaTeX template for academic CVs GitHub - oguzbilgic/posse: A web UI for Anthropic Managed Agents. GitHub - sshiraz/depsly: Dependency risk analysis tool for npm packages ABI Add safari/agent-harness — Safari browser automation via safari-mcp by achiya-automation · Pull Request #212 · HKUDS/CLI-Anything GitHub - Halfblood-Prince/trustcheck: Verify PyPI package attestations and improve Python supply-chain security GitHub - oguzbilgic/kern-ai: Agents that do the work and show it. GitHub - bruits/satteri: High-performance Markdown and MDX processing for the JavaScript ecosystem GitHub - tylergibbs1/feedstock: High-performance web crawler and scraper for TypeScript, powered by Bun and Playwright GitHub - Grimm67123/grimmbot: The self-improving sandboxed and open-source AI agent. With persistent memory and scheduling. GitHub - whitevanillaskies/whitebloom: Local whiteboard that blooms. GitHub - hwdsl2/docker-whisper: Docker image for a self-hosted Whisper speech-to-text server with speaker diarization and OpenAI-compatible transcription and translation APIs. Powered by faster-whisper. Supports all Whisper models, NVIDIA GPU (CUDA) acceleration, JSON/SRT/VTT output, SSE streaming, offline mode, and multi-arch (amd64, arm64). GitHub - yisding/reviewwiggum GitHub - MarwanAlsoltany/serrors: Structured errors for Go: sentinel hierarchies, typed data, custom formatting, and slog integration. GitHub - soatok/age-php GitHub - Luthiraa/markitme GitHub - stagas/rtdiff: realtime git diff gui and AI-assisted commits GitHub - tombedor/excalicharts GitHub - wh1le/excalidraw-edit: Open and edit .excalidraw files from the terminal. Offline, auto-saves to disk. MalExt Sentry - Malicious Extension Scanner - Chrome 应用商店 GitHub - syi0808/asciianimesvg: Generate animated ASCII art SVGs from text. CLI, Rust library, WASM, and web editor. GitHub - zaina-ml/ml_forge: A visual-based graph node editor for training computer vision models. GitHub - anakin87/llm-rl-environments-lil-course: 🌱 A little course on Reinforcement Learning Environments for evaluating and training Language Models GitHub - takaakit/superpowers-uml: Superpowers-UML modifies Superpowers to ensure a software development workflow in which AI agents design through UML modeling. AdriByte Studio - Sviluppo Web e Soluzioni Digitali GitHub - chouligi/angel-copilot: Your personalized Angel Investment Advisor Show HN: MoodSense AI (ML and FastAPI and Gradio, Deployed on Hugging Face) Moodsense Ai - a Hugging Face Space by aman179102 GitHub - agenteractai/lodmem: Level Of Detail Context Management for Agents GitHub - ostefani/subnetlens: A fast, concurrent network scanner with a TUI and plain-text CLI, built in Go. It discovers live hosts on your network, scans their open ports, resolves hostnames, and fingerprints operating systems—delivered. Cyber Pulse: Agentic Intel - Apps on Google Play Whisper API: Self-Hostable Speech to Text Transcription The Agent-Web Protocol Stack: A Research Thesis GitHub - msmarkgu/RelayFreeLLM: A restful API designed to route user prompts to various AI model providers. Show HN: Provepy – A Python decorator that proves your code using Lean and LLMs Show HN: Pardonned.com – A searchable database of US Pardons GitHub - patrickdappollonio/dux: Dux is a terminal UI that lets you run multiple AI coding agents side by side, each in its own git worktree, with full companion terminals, macros, commit generation, and a command palette that knows more tricks than you do. kMC Crystal Simulator Show HN: HyperFlow – A self-improving agent framework built on LangGraph GitHub - stef41/vibescore: 🎵 Grade your vibe-coded project. One command, instant letter grade across security, quality, dependencies, and testing. GitHub - stef41/lmscan: 🔍 Detect AI-generated text and fingerprint which LLM wrote it. Open-source GPTZero alternative. Zero dependencies, works offline. imgur.com GitHub - visionscaper/collabmem: Enabling long-term collaboration with Agentic AI - building up episodic and world model memory over time with in-context awareness 在 Steam 上购买 FriedrichAI: Offline AI 立省 10% GitHub - atripati/ark: AI Runtime Kernel — a context operating system for AI agents. Eliminates tool bloat, loads only what’s needed, and gives LLMs their reasoning space back. GitHub - nowork-studio/toprank: Open-source Claude Code skills for SEO, SEM, Google Ads GitHub - tacomanator/sash: Lightweight macOS menu bar app for reliably cycling through windows of the current application. Appents | Social Media Management for Product-First Teams GitHub - pnhoang/youtube-spam-blocker: Automatically detects and hides spam messages in YouTube Live chat. Set rate limits, keyword filters, and block repeat offenders. GitHub - decisionnode/DecisionNode: CLI + Local MCP - A shared structured memory store across Claude Code, Cursor, Windsurf, Antigravity, and every MCP client. Semantically queryable. GitHub - AvaCodeSolutions/django-email-learning: An open source Django app for creating email-based learning platforms with IMAP integration and React frontend components. The $100K Gap in Kubernetes Security Tooling Function Calling Harness: From 6.75% to 100%
GitHub - vishal-dehurdle/state-harness: Runtime safety net for LLM agents. Detects token spirals, kills doomed tasks early, tells you exactly why. Rust core, Python SDK. pip install state-harness
visha1v · 2026-06-10 · via Hacker News: Show HN

PyPI Downloads Python Rust Core License: Split BSL/Apache DOI

Lyapunov-stability monitor for multi-turn LLM agents. Detects token spirals, classifies failure patterns, and tells you why a task failed — no extra LLM calls.

from state_harness import GrowthRatioGuard, FailureReport

guard = GrowthRatioGuard(token_budget=50_000)

with guard:
    for turn in agent_loop:
        result = llm.invoke(turn.prompt)
        guard.record_step(tokens_used=result.usage.total_tokens)

# What went wrong? (zero-cost, no LLM calls)
report = FailureReport.from_guard(guard)
print(report)
⚠️  STABILITY TRIPPED at turn 12

Pattern: Context Accumulation Spiral (confidence: 92%)
  • Last 5 turns all exceeded 1.5× baseline (4/4 were accelerating).
  • Peak growth ratio: 5.2× baseline.
  • Without intervention, projected cost was $0.0396 (actual: $0.0039).

Energy: ▁▁▁▁▁▂▂▃▄▆█
  Baseline: 1050 tokens/turn
  Peak ratio: 5.2× baseline

Cost: $0.0039 (saved ~$0.0357 by tripping early)

Suggested actions:
  🔴 1. Enable RG history compression in your agent loop.
     → Compressing older messages reduces prompt tokens by 40-60%.
  🟡 2. Lower the growth ratio threshold to 1.8×.
     → A lower threshold would have caught it earlier.
  🟢 3. Add a sliding-window context strategy.
     → Send only the last N messages plus a summary of earlier ones.

Why this exists

Production multi-agent systems fail at rates of 41–87% (Kore.ai 2026). When an agent spirals — replaying full context, retrying a broken tool, drifting off-task — a budget cap will kill it, but tells you nothing about why.

State-harness monitors token consumption relative to a warmup baseline via a Lyapunov energy function. When the growth ratio exceeds a threshold for W consecutive steps, it trips and classifies the failure pattern (context spiral, retry storm, policy drift) with fix suggestions — from the energy trajectory alone, no LLM calls.

pip install state-harness and wrap your agent loop.

What it catches

Pattern Signal Example
Context Spiral Token growth accelerating beyond baseline Agent replaying full history each turn
Retry Storm Low-variance repeated calls Tool failing, agent retrying identically
Policy Drift VSA similarity score dropping Agent going off-topic mid-conversation
Early Explosion Token spike in first 3 turns Oversized system prompt or tool response
Budget Exhaustion Cumulative spend hits ceiling Complex task, not necessarily broken

Scope and limitations

State-harness does not improve resolve rates — a naive budget cap achieves comparable task success (multi-trial results below). The value is:

  1. Failure diagnostics — classified failure patterns with actionable fixes, not just "budget exceeded." No extra LLM calls.
  2. Compute efficiency on long loops — 38.6% fewer search nodes and 30% less wall time on SWE-bench by terminating dead-end branches early.

Validated across 3,175 runs (4 benchmarks, 5-condition ablation, multi-trial with bootstrap CIs). Zero false positives across 7 models incl. 4 local via Ollama. Details in Benchmarks.

When to use this

  • Search-tree agents (MCTS, beam search) — per-branch caps look fine in isolation; tree-level cost explosion is silent.
  • Platform teams at scale — failure classification at the edge, exported as OpenTelemetry attributes.
  • Benchmarking — the ~4–5% nondeterminism floor means single-run deltas <8% are noise.

Not needed for chatbots, RAG, single-turn apps, or ReAct loops with <10 turns — max_iterations + budget cap suffice.


Installation

pip install state-harness

Python ≥ 3.10. Pre-built wheels for Linux, macOS, Windows (x86_64 + ARM64). No Rust toolchain needed.

From source (for development)

git clone https://github.com/vishal-dehurdle/state-harness.git
cd state-harness

python -m venv .venv && source .venv/bin/activate

# Install Rust (if not already installed)
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

pip install maturin
maturin develop --release

# Run tests
pip install pytest
pytest tests/

Quickstart

Basic: GrowthRatioGuard (recommended)

GrowthRatioGuard normalizes token usage against a baseline — trips only on disproportionate growth, not natural context-window accumulation.

from state_harness import GrowthRatioGuard, StabilityViolation

guard = GrowthRatioGuard(
    token_budget=100_000,     # hard ceiling
    ratio_threshold=2.0,      # trip when turn is 2× the baseline
    window=3,                 # 3 consecutive escalating turns to trip
    budget_gate=8_000,        # don't trip until 8K tokens spent
)

with guard:
    for turn in agent_loop:
        try:
            result = llm.invoke(turn.prompt)
            guard.record_step(
                tokens_used=result.usage.total_tokens,
                errors=0,
            )
        except StabilityViolation as e:
            print(f"Agent killed: {e}")
            break

print(f"Total cost: {guard.total_tokens} tokens")
print(f"Baseline: {guard.baseline} tokens/turn")
print(f"Peak ratio: {guard.current_ratio}×")

Failure Diagnostics

After any execution (tripped or not):

from state_harness import FailureReport

report = FailureReport.from_guard(guard, model="gemini-2.5-flash")

# Human-readable terminal output
print(report)

# Structured dict for logging / dashboards
import json
print(json.dumps(report.to_dict(), indent=2))

Classifies the failure pattern, provides evidence, estimates cost, and suggests fixes — no LLM calls.

Classic: BoundaryGuard

For lower-level control using raw token counts (no normalization):

from state_harness import BoundaryGuard

with BoundaryGuard(token_budget=100_000, lambda_=1.0, window=5) as guard:
    for turn in agent_loop:
        result = llm.invoke(turn.prompt)
        guard.record_step(
            tokens_used=result.usage.total_tokens,
            errors=0,
            tool_name="search",
        )

Decorator: @boundary_guard

from state_harness import boundary_guard

@boundary_guard(
    token_budget=50_000,
    token_counter=lambda r: r.usage.total_tokens,
)
def agent_step(prompt: str):
    return llm.invoke(prompt)

Framework Integration

LangGraph (recommended)

from langgraph.prebuilt import create_react_agent
from state_harness.adapters import monitor_graph

agent = create_react_agent(model, tools=[search, calculate])
safe = monitor_graph(agent, token_budget=100_000)

result = safe.invoke({"messages": [("user", "Fix the login bug")]})

# After execution — always available:
print(safe.total_tokens)  # cumulative usage
print(safe.tripped)       # did stability trip?
print(safe.report)        # full FailureReport with pattern + suggestions

For streaming:

for chunk in safe.stream({"messages": [("user", "Refactor this module")]}):
    print(chunk)

With a trip callback (e.g., for Slack alerts):

safe = monitor_graph(
    agent,
    token_budget=100_000,
    on_trip=lambda report: slack.send(f"Agent tripped: {report.pattern}"),
)
Advanced: per-tool wrapping with LangGraphMiddleware
from state_harness import BoundaryGuard
from state_harness.adapters import LangGraphMiddleware

guard = BoundaryGuard(token_budget=150_000)
middleware = LangGraphMiddleware(guard)

@middleware.wrap_tool
def search_database(query: str):
    return db.search(query)

with guard:
    result = agent.invoke({"messages": [...]})

CrewAI

from crewai import Agent, Task, Crew
from state_harness.adapters import CrewAICallback

callback = CrewAICallback(token_budget=200_000)

crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    step_callback=callback.step_callback,
    task_callback=callback.task_callback,
)

result = crew.kickoff()
print(callback.report)  # FailureReport
callback.close()

Vanilla Python Hooks

from state_harness import BoundaryGuard
from state_harness.adapters import VanillaHook

guard = BoundaryGuard(token_budget=50_000)
hook = VanillaHook(guard)

with guard:
    for step in agent_loop:
        hook.before_call(tool_name="search")
        result = execute_tool(step)
        hook.after_call(tokens_used=result.tokens)

CLI

# Simulate a token trajectory — see what the guard would do
state-harness simulate 1000 1200 1500 2000 3000 5000 8000 --budget 50000

# Analyze a saved report
state-harness analyze report.json
state-harness analyze report.json --json    # JSON output
state-harness analyze report.json --otel    # OpenTelemetry attributes

# Batch analyze all reports in a directory
state-harness batch --dir ./reports/ --output results.csv

Structured Output

FailureReport supports multiple output formats:

report = FailureReport.from_guard(guard)

# JSON (for logging, APIs, storage)
report.to_json()            # pretty-printed
report.to_json(indent=None) # compact, single line

# CSV (for batch analysis of 1000s of runs)
with open("results.csv", "w") as f:
    f.write(FailureReport.csv_header() + "\n")
    for r in reports:
        f.write(r.to_csv_row() + "\n")

# OpenTelemetry (for Datadog, Grafana, Honeycomb)
from opentelemetry import trace
span = trace.get_current_span()
span.set_attributes(report.to_otel_attributes())
# Adds: state_harness.pattern, state_harness.confidence, etc.

Architecture

Three mechanisms, implemented in Rust (via PyO3):

graph TD
    A["Agent Loop"] --> B["GrowthRatioGuard\n(Python SDK)"]
    B --> |"Normalizes tokens → growth ratio\nWarmup baseline · Budget gate"| C{" "}
    C --> D["Lyapunov Monitor\nV(k) = S + λθ\nΔV ≥ 0?"]
    C --> E["RG Decimator\nTF-IDF\nCompression"]
    C --> F["Holographic Engine\n(VSA)\nDrift Detection"]
    
    style D fill:#1a1a1a,stroke:#555,color:#e8e8e8
    style E fill:#1a1a1a,stroke:#555,color:#e8e8e8
    style F fill:#1a1a1a,stroke:#555,color:#e8e8e8
    style B fill:#0d1117,stroke:#30363d,color:#e6edf3
Loading
Component Purpose Speed
Lyapunov Monitor Tracks energy derivative ΔV(k). Trips when ΔV ≥ 0 for W consecutive steps. ~1μs/step
RG Decimator RG-inspired decimation of conversation history (TF-IDF scoring). Retains structurally important messages. ~100µs/compress
Holographic Engine VSA-based policy drift detection. Binds domain invariants to high-dimensional vectors. ~10μs/check

Benchmarks

5-condition ablation across 4 benchmarks (3,175 total runs). Full methodology in the research paper.

Ablation Conditions

Condition Lyapunov RG Decimation VSA Dual-Gate Description
A. Baseline Unmonitored agent
B. Lyapunov-only Energy monitoring, no intervention
C. Lyapunov+RG + history compression on violation
D. Full-stack + policy drift gating
E. Naive Cap Hard budget cap (control)

Summary

Benchmark Runs Stability Trips Cost Savings (D vs A) Resolve-Rate Δ Diagnostics
MINT (reasoning + coding) 1,136 0 ~0% −0.7pp (noise) N/A (no trips)
τ³-bench (customer service) 750 0 8.1% within ±12pp nondeterminism N/A (no trips)
SWE-bench Verified (coding) 333 + 148 ~38% 38.6% (nodes) −3.6pp (within ±4–5% noise) Pattern classification
Custom Local (4 models) 240 3 (true pos.) 15.2% 0pp Pattern classification
MINT Local (Qwen3:4B) 568 0 ~0% +1.8pp N/A (no trips)

Resolve-rate deltas fall within LLM nondeterminism (~4–5% stdev). No trips on short/medium loops (1,886 runs). Savings concentrate on long-loop search trees.

SWE-bench Verified (central result)

37 Django instances, SWE-bench Verified. Agent: moatless-tools SearchTree, 50-node budget. Model: Gemini 2.5 Flash.

Single-trial ablation (148 runs)

Condition Resolved Rate Total Nodes Wall Time Nodes/Resolve
A. Baseline 15 / 37 40.5% 945 80 min 63.0
B. Lyapunov 16 / 37 43.2% 620 69 min 38.8
D. Full-stack 14 / 37 37.8% 580 56 min 41.4
E. Naive Cap 21 / 37 56.8% 876 77 min 41.7

Note: Single-trial resolve rates have ~±8pp standard error. E's apparent 56.8% is not statistically significant vs A's 40.5%. Multi-trial results below confirm this.

Full-stack monitoring: 38.6% fewer nodes (945 → 580), 30% less wall time (80 → 56 min). Baseline had 7 tasks burning the full 50-node budget (all failed); with monitoring, zero hit ceiling. Lyapunov alone (Condition B, ~5 lines of code) delivers ~90% of the savings.

Ablation — each mechanism contributes independently:

Layer Added Compute (nodes) Δ vs Baseline Cumulative Reduction
A. No monitoring 945
B. + Lyapunov 620 −325 34.4%
D. + RG + VSA 580 −40 38.6%

Lyapunov alone delivers ~90% of the benefit. RG and VSA add incremental value.

Multi-trial validation (333 runs)

3 trials per condition (A, D, E) across all 37 instances — 333 total runs. 12 runs stuck in Docker (28+ min), counted as failures:

Condition Trial 1 Trial 2 Trial 3 Mean ± σ
A. Baseline 18/37 (48.6%) 16/37 (43.2%) 15/37 (40.5%) 44.1% ± 4.1%
D. Full-stack 15/37 (40.5%) 16/37 (43.2%) 14/37 (37.8%) 40.5% ± 2.7%
E. Naive Cap 19/37 (51.4%) 15/37 (40.5%) 17/37 (45.9%) 45.9% ± 5.4%

Cross-condition variance (2.9%) ≤ within-condition nondeterminism (4.1%). All differences fall within the noise band.

The ~4% within-condition stdev converges with τ³-bench (±4.6%), establishing a ~4–5% nondeterminism floor for Gemini 2.5 Flash on code tasks. Single-run deltas <8% are unreliable.

Bootstrap CIs (10,000 resamples) and Welch's t-tests: A−D = +3.6pp [−0.9, +8.1], p ≈ 0.17; A−E = −1.8pp [−8.1, +4.5], p ≈ 0.68; D−E = −5.4pp [−10.8, 0.0], p ≈ 0.09. Full analysis in paper §7.3.1.

τ³-bench Airline (non-invasiveness confirmation)

50 tasks × 3 trials × 5 conditions = 750 total runs. Agent handles airline reservations via tool calls. Model: Gemini 2.5 Flash. Concurrency=1.

Condition Trial Pass Rate Task Pass (maj) Rate Cost Cost Δ
A. Baseline 99/150 66.0% 35/50 70.0% $2.47
B. Lyapunov-only 83/150 55.3% 28/50 56.0% $2.42 −2.0%
C. Lyapunov+RG 79/150 52.7% 26/50 52.0% $1.69 −31.8%
D. Full-stack 86/150 57.3% 30/50 60.0% $2.28 −8.1%
E. Naive Cap 81/150 54.0% 26/50 52.0% $2.33 −5.7%

Key findings:

  • Zero stability trips across 750 runs. All airline tasks classified as stable; no interventions.
  • Pass-rate variance is nondeterminism. Naive cap (E, zero monitoring) drops −16pp from baseline — worse than full-stack (D, −10pp). The ~10–16pp spread is intrinsic variance.
  • 25% of tasks flip pass/fail within the same condition across trials (~±12pp nondeterminism floor).
  • 8.1% cost savings from passive monitoring (zero interventions).

MINT (non-invasiveness validation)

284 tasks × 4 conditions = 1,136 total runs across GSM8K (48), MATH (100), HumanEval (45), MBPP (91). Agent uses up to 5 turns per task.

Condition GSM8K MATH Total Tokens
A. Baseline 91.7% 39.0% 29.2% 1,909,582
B. Lyapunov 91.7% 41.0% 29.9% 1,904,421
C. Lyapunov+RG 89.6% 37.0% 28.2% 1,910,926
D. Full-stack 87.5% 39.0% 28.5% 1,949,708

Zero stability violations across 1,136 runs. Token usage invariant (<2% overhead).

Failed tasks cost disproportionately more:

Task Success Avg Failure Avg Ratio
GSM8K 2,613 tok 8,857 tok 3.4×
MATH 5,154 tok 8,188 tok 1.6×

HumanEval and MBPP show 0% across all conditions — a MINT framework limitation in code execution evaluation, consistent across conditions (harness does not introduce new failure modes).

Local Model Validation (edge deployment)

20 custom tasks (5 easy, 10 medium, 5 hard) × 4 models × 3 conditions = 240 runs. Hardware: Apple M4 MacBook Pro, 16 GB RAM, Ollama local inference.

Model Size Baseline Harness Naive Cap Token Savings FP
Llama 3.2:3B 2.0 GB 45% 45% 60% 1.2% 0
Phi-4-Mini 2.5 GB 30% 30% 40% 20.7% 0
Qwen3:4B 2.5 GB 30% 30% 40% 0.9% 0
Gemma4:E4B 9.6 GB 35% 35% 70% 37.9% 0

Key findings:

  • Zero false positives across 80 harness runs — 4 model families, 3 difficulty tiers. Growth-ratio generalizes without threshold retuning.
  • Small-model self-sabotage: Naive cap beats baseline by +17.5pp avg (+12.5pp median). Small models solve early turns correctly, then destroy solutions in later turns. Strongest on Gemma4:E4B (+35pp).
  • Model-family behavioral signatures:
    • Llama 3.2:3B: Classic spirals (ratios: 2.3×, 5.9×, 7.6×) — 3 true-positive trips
    • Phi-4-Mini: Spike-and-recover — 20.7% passive savings
    • Qwen3:4B: 255K tokens but flat ratios (≤1.06×) — stable despite 3× volume
    • Gemma4:E4B: Decreasing ratios — 37.9% passive savings, zero trips

Deploying ≤4B models via Ollama? State-harness works out of the box (zero false positives). The self-sabotage finding suggests adding a turn limit (2–3 turns) for open-ended code generation.

MINT on Qwen3:4B (568 runs)

Task Harness (max=5) Naive Cap (max=2) Δ
GSM8K 37.5% 27.1% +10.4pp
MATH 0.0% 0.0%
HumanEval 11.1% 11.1%
MBPP 14.3% 14.3%
Total 12.7% 10.9% +1.8pp

Zero interventions across 284 tasks. With max 5 turns and W=3, the monitor cannot trigger within available post-warmup turns — a structural guarantee.

Reproducing the benchmarks

Full reproduction steps (all three benchmarks)
# 1. Clone repos
git clone https://github.com/vishal-dehurdle/state-harness.git
git clone https://github.com/sierra-research/tau-bench.git tau3-bench

# 2. Install state-harness
cd state-harness
python -m venv .venv && source .venv/bin/activate
pip install maturin && maturin develop --release

# 3. Install τ³-bench (with state-harness agent)
cd ../tau3-bench
uv sync
cp ../state-harness/tau3_integration/harness_agent.py src/tau2/agent/
cp ../state-harness/tau3_integration/naive_cap_agent.py src/tau2/agent/

# 4. Configure Vertex AI
export GOOGLE_CLOUD_PROJECT=your-project-id
export VERTEXAI_LOCATION=asia-south1

# 5. Run τ³ 5-phase benchmark
bash benchmarks/tau3/run_5phase_airline.sh

# 6. Run SWE-bench (requires Docker images)
bash benchmarks/swe_bench/run_benchmark.sh
bash benchmarks/swe_bench/run_benchmark_dbe.sh

# 7. Run MINT
bash benchmarks/mint/run_mint_fullstack.sh

Ablation conditions are controlled via environment variables:

Variable Values Effect
HARNESS_RG on / off Enable/disable RG history compression
HARNESS_VSA on / off Enable/disable VSA policy drift detection
HARNESS_RATIO_THRESHOLD float (e.g., 2.0) Override growth ratio threshold
HARNESS_BUDGET_GATE int (e.g., 8000) Override minimum spend before trip

See benchmarks/ for setup, configs, and reproduction instructions.

Future evaluations

  • Multi-trial SWE-bench — 333 runs (3 trials × 3 conditions × 37 instances) confirming non-invasiveness within ±4% noise band
  • Local model validation — 240 runs across 4 open-weight models (Llama, Phi, Qwen, Gemma) + 568 MINT runs on Qwen3:4B
  • Terminal-Bench — Terminal-based agent tasks; command-line tool loops where spirals manifest as repeated failed commands
  • SWE-bench Pro — Harder, contamination-resistant variant of SWE-bench
  • Cross-model validation — 7 models total: GPT-4o-mini, Claude Haiku 4.5, Gemini 2.5 Flash + Llama 3.2:3B, Phi-4-Mini, Qwen3:4B, Gemma4:E4B

Known limitations

  1. 37 SWE-bench instances — A larger sample would improve statistical power (n=3 trials gives limited degrees of freedom for t-tests).
  2. No causal intervention — The harness currently kills spiraling tasks. Redirect/repair is on the roadmap.
  3. Physics-inspired, not physics-equivalent — Terms like "Renormalization Group" and "Lyapunov stability" are used as structural inspirations. The mathematical mapping is analogical, not isomorphic.
  4. Custom benchmark scale — The 20-task local battery is smaller than standard benchmarks. The self-sabotage finding (mean +17.5pp, median +12.5pp) is consistent across 4 models but requires larger-scale replication.

Configuration Guide

Parameter Default Description
token_budget 100,000 Hard ceiling on cumulative tokens
ratio_threshold 2.0 Growth ratio above which a turn counts as "escalating" (domain-tuned: airline=2.0, retail=2.5, telecom=2.0)
window 3 Consecutive escalating turns before circuit breaker trips
warmup_turns 3 Turns used to establish baseline (no monitoring during warmup)
budget_gate 8,000 Minimum cumulative tokens before the monitor can trip (retail: 12,000)
lambda_ 1.0 Error weighting in the Lyapunov energy function

Environment variable overrides (highest precedence, for threshold sweeps):

Env Var Description
HARNESS_RATIO_THRESHOLD Override ratio_threshold (e.g., 2.5)
HARNESS_BUDGET_GATE Override budget_gate (e.g., 12000)

Tuning tips:

  • More aggressive (catch spirals earlier): ratio_threshold=1.8, window=2
  • More conservative (fewer false positives): ratio_threshold=2.5, window=3
  • High-value tasks: Increase budget_gate to 20K+ to let expensive tasks run longer
  • Complex domains (retail, multi-tool): Start with ratio_threshold=2.5

Theoretical Foundations

  • Lyapunov stability: V(k) = S(k) + λθ(k) models token consumption as a dynamical system. ΔV ≥ 0 for W consecutive steps → unstable.
  • Renormalization Group (RG): Message compression via coarse-graining — eliminates high-frequency noise, preserves scale-invariant task objectives.
  • Vector Symbolic Architecture (VSA): Domain policies bound to high-dimensional bipolar vectors (10,000-d, i8), enabling constant-time drift detection outside the LLM context window.

Research

Implements the framework from:

Empirical Lyapunov Stability: Growth-Ratio Energy Functions as Leading Indicators of Agent Task Failure Vishal Verma, 2026 DOI Read the full paper →

Full ablation, multi-trial validation, local-model results, and failure taxonomy. Key results reproduced in Benchmarks above.

Citation

If you use this library or refer to these findings in your research, please cite the preprint:

@misc{verma2026empirical,
  author       = {Verma, Vishal},
  title        = {Empirical Lyapunov Stability: Growth-Ratio Energy Functions as Leading Indicators of Agent Task Failure},
  month        = jun,
  year         = 2026,
  publisher    = {Zenodo},
  version      = {1.0.0},
  doi          = {10.5281/zenodo.20722987},
  url          = {https://doi.org/10.5281/zenodo.20722987}
}

Based on the theoretical framework from:

The Fluid Dynamics of Multi-Agent AI: Resolving d'Alembert's Paradox of Generative Workflows Vishal Verma, 2026 Read →


Contributing

See CONTRIBUTING.md for dev setup, code style, and PR guidelines.


Roadmap

  • Adaptive threshold — Auto-tune τ based on task complexity signal from early turns
  • Causal intervention — Instead of killing spiraling tasks, redirect them (prompt injection, tool restriction)
  • Streaming support — Token-level monitoring for streaming LLM responses
  • Multi-model validation — 7 models validated: GPT-4o-mini, Claude Haiku 4.5, Gemini 2.5 Flash + 4 local models via Ollama
  • Dashboard / observability — Optional lightweight UI for monitoring energy trajectories in real-time

Security

See SECURITY.md. Do not open public issues for security reports.


License

Split-core licensing:

Component License Notes
Rust Core (src/) BSL 1.1 Free for non-commercial + ARR < $1M. Converts to Apache 2.0 on May 26, 2030.
Python SDK (python/) Apache 2.0 Fully permissive.

See LICENSE.md for full details.