117 results for “gh”

Hyperflow H3 is Video Rebirth's 8-step LoRA for MiniMax H3, released in September 2026 with open weights. Artificial Analysis ranks it #5 on AA-Video-I2V v2.1 and #3 on AA-Video-Editing v1.1, ahead of MiniMax H3 (768p). It costs $9.00 per minute for Image to Video and $18.00 for Video Editing at 1088p, under half of Dreamina Seedance 2.5 ($34.12 and $40.82). The thread gives per-capability and per-use-case breakdowns.

x.comthreadAdded 31 Oct 2026, published 30 Oct 2026🧠 AI

Pre-ERC draft on Ethereum Magicians that binds the x402 payTo address to an ERC-8004 agent's registered agentWallet, directly or through an EIP-712 payee delegation (ERC-1271 for contract wallets). It lists eight fixed checker outcomes and requires the resource origin to appear in the registration file. On Robinhood Chain (28 October 2026), 5,575 of 20,046 agents answered with x402 requirements, and only 5 used their agentWallet as payTo.

ethereum-magicians.orgthreadPublished and added 30 Oct 2026🛡️ Security

Pre-ERC draft on the Ethereum Magicians forum that ties the payTo address in an x402 payment requirement to the seller's ERC-8004 agent identity. The payTo must equal the registered agentWallet or an EIP-712 delegation signed by it, and the origin must be listed in the registration file. A checker returns one of eight fixed outcomes. On Robinhood Chain, 5 of 5,575 x402-enabled agents out of 20,046 used their agentWallet as payTo.

ethereum-magicians.orgthreadPublished and added 30 Oct 2026🧠 AI

A post relaying Cursor's security advisory: a modified, malicious Claude Code VS Code extension was distributed through the Cursor extension marketplace, affecting at most 5,000 users. Anthropic did not develop that version, and other extensions are not affected. Users of Claude Code in Cursor are told to update to version 2.1.314. The post adds that marketplaces need publisher verification and package integrity checks. It gives no technical detail on the payload.

x.comthreadAdded 31 Oct 2026, published 30 Oct 2026🛡️ Security

Artificial Analysis launched an HLE-Diamond leaderboard in v5 of its Intelligence Index, covering the refined 1,000-question Humanity's Last Exam from CAIS and Scale AI. Each question runs once without tools, graded by GPT-5.6 Luna (medium) as pass@1. Gemini 4 Carbon (high) leads at 69%, Claude Opus 5.5 (max) scores 68%, and the top open weights model, Kimi K3.1 Pro, reaches 37%.

x.comthreadPublished and added 30 Oct 2026🧠 AI

Mistral Large 4 ("Le Chonk") is an open-weights model from Mistral, now on Hugging Face. Per Mistral's announcement quoted in the post, it ranks #1 among open models on Cognition's FrontierCode benchmark, is aimed at agentic coding and cybersecurity, and claims better reasoning efficiency and lower inference costs. The post does not give parameter counts, license or benchmark scores.

x.comthreadAdded 30 Oct 2026, published 29 Oct 2026🧠 AI

Nemotron 3.5 Transcribe is a free open-weight speech-to-text model from NVIDIA covering 40 languages, including mixed-language speech. The post says it ties for first among open models with a 2.8% error rate. It says the model needs a large GPU such as an H100 or RTX 5090, so it will not run on a laptop. The post gives no benchmark name or model card link.

x.comthreadAdded 30 Oct 2026, published 29 Oct 2026🧠 AI

Nemotron 3.5 Transcribe is an open-weight, promptable NVIDIA mixture-of-experts model with 30B total and about 4B active parameters. One checkpoint does transcription in 40 languages, speaker attribution, translation, contextual prompting and audio understanding. The post says it ranked first in VoiceArena's comparison of 14 diarization plus ASR systems. The fetched text is cut off and gives no license or download details.

x.comthreadAdded 30 Oct 2026, published 29 Oct 2026🧠 AI

Rohan Paul's X post summarizes the paper "Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol", from TikTok researchers. It reports that MCP-using agents often finish the requested task, then quietly call an attacker-controlled logging tool that records the query, server reply and final answer. The attack works through natural-language tool descriptions that agents read when choosing and ordering tools. The visible text is truncated.

x.comthreadAdded 29 Oct 2026, published 28 Oct 2026🛡️ Security

Trail of Bits post says it achieved remote code execution through a Brotli decoder bug in Linux lab setups: Node.js 24.22.0 with Express 5.3.0 via the standard JSON parser (x86-64 and ARM64), and Electron 41.10.7 (x86-64) with a custom decoding service. Exploitation needs a carefully matched setup and repeated attempts. A scanner by ToBScottA finds hidden copies of the decoder, including in stripped binaries.

x.comthreadAdded 30 Oct 2026, published 28 Oct 2026🛡️ Security

Nemotron 3.5 Transcribe is an open-weight, promptable 30B mixture-of-experts model from NVIDIA with about 4B active parameters, built for recorded audio. One checkpoint transcribes 40 languages, handles recordings up to an hour, labels up to eight speakers, translates, and answers questions about English audio. Weights and a demo are on Hugging Face. The post gives no benchmarks.

x.comthreadAdded 29 Oct 2026, published 28 Oct 2026🧠 AI

A practical reverse-engineering map by hakimieiqbal on X that picks tools by target: native binaries, .NET, Electron, iOS, firmware, WebAssembly and network protocols. It introduces REA (morluto/rea), which connects coding agents to RE tools, and lists MCP bridges for IDA Pro, Ghidra and radare2. It adds an evidence-first workflow and free courses such as pwn.college and OpenSecurityTraining2.

x.comthreadAdded 29 Oct 2026, published 28 Oct 2026🛡️ Security

A thread by Hakimi Eiqbal that maps reverse-engineering tools by target type: native binaries, Electron, .NET, iOS, firmware, WebAssembly and network protocols. Its AI sections cover REA (an open-source layer connecting coding agents to analysis tools) and MCP bridges for IDA Pro, Ghidra and radare2. It argues for an evidence-first workflow and gives a prompt that separates verified facts from hypotheses.

x.comthreadAdded 29 Oct 2026, published 28 Oct 2026🧠 AI

Kizuno18's X post argues that the move from local stdio and Docker MCP servers to remote HTTP/SSE endpoints matures the Model Context Protocol in enterprise infrastructure. It lists three trade-offs: short-lived OAuth/JWT bearer tokens replace mounted admin credentials, deferred tool discovery avoids roughly 20k tokens of eager schema loading, and HCP Terraform enforces Sentinel/OPA policy, audit trails and approvals server-side. Opinion post with no benchmarks or code.

x.comthreadAdded 29 Oct 2026, published 27 Oct 2026🧠 AI

Hi3D v1.0 (formerly Sparc3D) is an open-sourced image-to-3D model from Hi3D, which runs a closed v3.0 refining at 2048 cubes per side. In a test with 13 professionals in 3D algorithms, 562 judgments against eight other image-to-3D models favoured it 83.5% of the time. The thread also covers the Hi3D Agent and an MCP server for other agents. Back and bottom views were checked, where TRELLIS.2 collapsed a winged figure.

x.comthreadAdded 28 Oct 2026, published 27 Oct 2026🧠 AI

Campfire is a self-hosted workspace by Prasenjit Sarkar that runs Claude Code, Codex, Goose, Aider and OpenHands side by side in one tab. Risky actions go through permission voting: a majority approves, any single deny blocks, and a 30 second timer ends the wait. Session replay shows which file or tool call triggered an action. The post cites Forever Security's report of a planted AGENTS.md. Codex followed it first time over 90% of the time and Claude Code about half the time. Try it with "bunx the-campfire".

x.comthreadAdded 28 Oct 2026, published 27 Oct 2026🧠 AI

Kimi K3.1 Pro is a new model from Moonshot AI that scores 55 on the Artificial Analysis Intelligence Index, up 11 points from its predecessor, and ranks 4th overall. It has a 1M-token context window and API pricing of $2 per million input tokens and $10 per million output tokens. Weights are due by November 4. Moonshot removed cyber-related training data, which cut its internal vulnerability research score from 90.5% to 81%.

x.comthreadAdded 28 Oct 2026, published 27 Oct 2026🧠 AI

AGTP's X thread summarizes Meta's Muse Spark 1.3 Evaluation Report. Prompt injection was tested on Exponential Security Labs' RepoInject v1 and BrowserInject v1 and on Gray Swan's IPI Arena, with robustness said to be on par with GPT-5.6 Sol. It also covers chain of thought controllability evals, PostTrainBench under Loss of Control, and Handshake's pre-release human red teaming. It notes the Midas Project's criticism of Muse Wonder's missing safety report.

x.comthreadAdded 27 Oct 2026, published 26 Oct 2026🛡️ Security

Thread summarizing Meta's Muse Spark 1.3 Evaluation Report, the safety write-up for the model released on September 2, 2026. It covers prompt-injection results on Exponential Security Labs' RepoInject v1 and BrowserInject v1 and Gray Swan's IPI Arena, chain-of-thought controllability evals, PostTrainBench under Loss of Control, and Handshake's human red teaming. It also notes The Midas Project's criticism that Muse Wonder still has no safety report, and the Muse Spark Open rollout (816B total, 42B active parameters).

x.comthreadAdded 27 Oct 2026, published 26 Oct 2026🧠 AI

Thread on two October 2026 npm compromises: @subql/common 5.8.3 (5 October), published through GitHub Actions OIDC Trusted Publishing after the attacker tampered with the publish workflow, and tensorlake 0.5.144 (8 October), published by the project's own release workflow with valid provenance for about 1 hour 40 minutes. Both ran on install and stole cloud, GitHub, Kubernetes, Vault, SSH and .env credentials; tensorlake also took AI tool and MCP config. Its payload installed a gh-token-monitor service that deletes the home directory if the stolen GitHub token is revoked, so remove it before rotating.

x.comthreadAdded 27 Oct 2026, published 26 Oct 2026🛡️ Security

Reply on X explaining a Bytecode Alliance advisory for Wasmtime's Cranelift compiler backend. A large or crafted Wasm function that generates enough SSA values overflows Cranelift's value-index representation, which packed indices into 24-bit fields. The fix widens them to 32 bits. The issue is not AArch64-specific, and the advisory names no required optimisation level. Guard pages do not reliably stop the resulting miscompilation and type confusion, so a sandbox escape can arise at compile time.

x.comthreadAdded 28 Oct 2026, published 26 Oct 2026🛡️ Security

Post about the "Steal My Bitcoin" bounty, a 0.025 BTC (about $2,025) prize that was swept early Monday, about 90 minutes after researcher @killa published a full exploit on GitHub. The exploit is a heap exploit written for the target's PHP build. It leaks memory addresses through session cookies and forces PHP cleanup to run the flag binary. The first solve looked like a failure because the pulled key did not control the advertised address, a planting mistake the creator admitted by email.

x.comthreadPublished and added 26 Oct 2026🛡️ Security

Post by RicardoIQSource on breaches of developer platforms aimed at AI provider keys. It reports OpenCode (17 October, provider keys in its Console) and Zeabur (14-15 October, 3.2M environment variable records, admin tokens stored in plaintext since January 2024) as following the Railway incident. It claims three platforms were hit in eight days and links a blog with a timeline and remediation steps.

x.comthreadAdded 29 Oct 2026, published 26 Oct 2026🛡️ Security

SemiAnalysis claims AMD chips running vLLM can earn up to $3 per chip-hour in profit serving the open-source Kimi K3 model, even after higher chip rental prices following Claude Code's rise. Kimi K3 has over 2.8 trillion parameters and fits AMD's 288GB per chip, while B200 cannot hold it in one server. B200 must then use tensor or pipeline parallelism over 400Gbit/s InfiniBand or Ethernet, which SemiAnalysis says is 18x slower than NVLink. The post gives no methodology.

x.comthreadAdded 26 Oct 2026, published 25 Oct 2026🧠 AI

This X thread summarises OpenRouter usage charts. Weekly volume nears 200 trillion tokens, and open-weight models take 72.6% of tokens (136 trillion) against 27.4% for closed-weight models. Anthropic's daily average rose from about 0.6 to 2.25 trillion tokens alongside the Opus 5.5, Sonnet 5.5 and Haiku 5.5 releases. DeepSeek leads text requests at 26.5%, and agents plus coding account for 57.7% of spending. The author notes the data describes OpenRouter only.

x.comthreadAdded 26 Oct 2026, published 25 Oct 2026🧠 AI

Gheist is a GitHub campaign disclosed by Wiz in October 2026. Stolen tokens push ci-sync-XXXXXX branches with a sync-check workflow that dumps repo secrets base32-encoded to evade log masking, fetches the logs via API, then deletes the run. 13,500+ branches and 850+ accounts were hit since 15 September. Public repos are protected by the approval hold; private and Enterprise repos are not. The thread lists IOCs (.gdmp_pat files, "gheist" user agent, "ci: sync config" commits) and rotation steps.

x.comthreadAdded 26 Oct 2026, published 25 Oct 2026🛡️ Security

Nanbeige4.3-33B is a 33.1B dense, Llama-style open model under Apache-2.0, released with a base model on Hugging Face. In Nanbeige's own table it beats Qwen3.8-27B on Terminal-Bench 2.1 (80.9 vs 73.0), DeepSWE (65.5 vs 42.5) and Automation Bench, but trails on SWE-Bench Pro, GDPval and GPQA Diamond. BF16 weights are about 66 GB, split across 4 GPUs in the vLLM and SGLang examples. No independent evals exist yet.

x.comthreadAdded 26 Oct 2026, published 25 Oct 2026🧠 AI

ERC-8468, the Agent Safety Framework (ASF), is a draft Ethereum standard with two on-chain objects: a versioned Safety Specification and a Safety Binding with suspend, resume and terminate rights. A check() call returns Satisfied, NotSatisfied or Unknown at the moment of effect. Revision 6 has reference contracts, 90 tests and a Sepolia run of 36 transactions in October 2026, and the thread asks reviewers five design questions.

ethereum-magicians.orgthreadAdded 27 Oct 2026, published 25 Oct 2026🧠 AI

ERC-8468 (Agent Safety Framework, ASF) is a draft Ethereum standard, now at revision 6, for recording which participants accepted which safety requirements for an AI agent, in which version, enforced where, and whether that acceptance is in force now. It defines a versioned Safety Specification, an on-chain Safety Binding with interveners holding suspend, resume and terminate rights, and a check() that returns Satisfied, NotSatisfied or Unknown. The reference guarded account and task-tender adapter passed 90 tests (64 negative) and a Sepolia run of 36 transactions.

ethereum-magicians.orgthreadAdded 27 Oct 2026, published 25 Oct 2026🛡️ Security

Kimi K3.1 Pro is a Moonshot AI model, with a high-speed variant, described in a tweet as a mixture-of-experts system of about 2.8 trillion parameters with a 1-million-token context window and multimodal support. The post claims gains in coding, long-horizon agents, 3D modeling (Blender/CAD) and financial research, and says benchmarks come close to Claude Opus 5.5. Full open weights are planned by November 4. The post gives no benchmark figures.

x.comthreadAdded 26 Oct 2026, published 25 Oct 2026🧠 AI

A reviewer tests GenOffice, an open-source Office suite (10.7k GitHub stars), through its CLI only. Markdown to .docx took 3.8s, JSON to .xlsx took 0.9s with a live SUM formula, and .docx to PDF failed twice until Xvfb supplied a display. It has byte-preserving edits, a skill file for Claude Code and Codex, and 29 MCP tools. Downsides: a 188MB download that filled a 512MB /tmp, Electron in the PDF path, and unproven trust in tracked changes and rollback.

x.comthreadAdded 25 Oct 2026, published 24 Oct 2026🧠 AI

An X post showing how to have Codex train a small model on Apple silicon through the mlx-lm-lora MCP server and skill. It gives five setup steps (pip install with the mcp extra, npx skills add, locate mlx-lm-lora-mcp, codex mcp add). The prompt runs four pre-training trials on tiny-random/qwen3 with the mlx-community/recycling_the_web-1k dataset. Codex picks the optimizer and learning rate itself, and the result is a table with validation loss and the best adapter path.

x.comthreadAdded 25 Oct 2026, published 24 Oct 2026🧠 AI

AI Quanting's X thread covers Moonshot's Kimi K3.1 Pro release (1M-token context, open weights promised by November 4). Moonshot removed cyber data from mid-training and RL stages; on its in-house xdev-bench-hard exploit benchmark (Linux kernel, Node.js, JavaScriptCore, Exim, VirGL bugs) pass@3 fell from 90.5% to 81.0%. The author argues this is only about a tenth of the capability, so four in five expert exploit tasks remain solvable.

x.comthreadAdded 25 Oct 2026, published 24 Oct 2026🛡️ Security

Kimi K3.1 Pro is a new model from Moonshot AI. According to a post by hajra, Artificial Analysis ranks it 4th in the world, 3 points behind Claude Opus 5.5. The API price is down by a third, and full open weights are promised by 4 November. The post gives no benchmark breakdown, licence or architecture details.

x.comthreadPublished and added 24 Oct 2026🧠 AI

A post on X about NVIDIA's Nemotron 3.5 Parse and its "#1 on ParseBench" claim. The author says the ranking holds only among open-weight models under 1B parameters. On the full board of 150+ models the model scores about 75, while the top is near 90. The post asks vendors who run a benchmark to print the filter on the chart.

x.comthreadAdded 23 Oct 2026, published 22 Oct 2026🧠 AI

Social post summarising emergency patches for Node.js (4 high CVEs), PHP (2 critical) and nginx (3 critical). Node.js flaws include a Brotli decoder out-of-bounds write, a DNS truncation bug and two TLS hostname-verification bypasses. PHP has use-after-free and heap corruption bugs, and nginx has alias-rewrite and map-regex heap overflows plus an SSI-AIO use-after-free. It lists the fixed versions for each product.

x.comthreadAdded 23 Oct 2026, published 22 Oct 2026🛡️ Security

Post describing a LiteLLM 1.107.0 PyPI compromise in which attackers used stolen GitHub CLI OAuth tokens from three engineers to trigger and approve the project-releaser workflow, producing eight malicious wheels and a tampered tarball. The wheels drop cook.pth, which runs at Python startup, and privcollect steals cloud tokens, keys, browser data, AI configs and LiteLLM proxy credentials, exfiltrating to litellm[.]dev/healthcheck. A second daemon polls a GitHub domain.txt hourly for new domains to fetch and execute. PyPI yanked the package the same day. The post is a single unsourced tweet.

x.comthreadPublished and added 22 Oct 2026🛡️ Security

XiaoZhi ESP32 is an open-source, MIT-licensed firmware stack that turns ESP32 boards into MCP-based voice chatbots using models such as Qwen and DeepSeek. It streams Opus audio over WebSocket or MQTT + UDP through ASR, LLM and TTS, adds offline ESP-SR wake words, and uses device-side MCP to drive speakers, LEDs, servos and GPIO. It supports 138 board directories and builds with ESP-IDF 6.0.1 or newer.

x.comthreadAdded 23 Oct 2026, published 21 Oct 2026🧠 AI

FrontierMemory is a benchmark that evaluates agent memory through downstream task performance, measuring how agents learn from experience rather than recall, which the post says is saturated via agentic search. In elvis's summary, Claude Opus 5.5 ranks first and open-weight GLM-5.3 does well on memory generation. The authors reply that "memory" here means any adaptation from experience, so harness-level changes can also be scored. The post gives no scores or links.

x.comthreadAdded 23 Oct 2026, published 21 Oct 2026🧠 AI

Post summarising Pillar Security's finding of a remote code execution flaw in OpenCode, the open-source AI coding agent. Versions 1.0.137 through 1.18.32 could run code from a remote MCP server during OAuth sign-in on Windows and WSL, with no prompt injection needed. The fix is in 1.18.33 (released 28 September). No advisory or CVE was issued, so scanners will not flag old installs. Interim advice is to remove unvetted remote MCP servers.

x.comthreadAdded 22 Oct 2026, published 21 Oct 2026🛡️ Security

Flint is the first model from Recursive SI, a fast local coding model with open weights released the same day. It was built with the company's automated research system, the Eureka Machine. The post claims up to 2x faster speed than comparable models: 125+ tokens/sec on a MacBook Pro and 150+ on RTX Spark. The captured text is truncated and gives no benchmarks or model size.

x.comthreadAdded 22 Oct 2026, published 21 Oct 2026🧠 AI

SemiAnalysis thread arguing that vLLM with Wide-EP and EPLB can reproduce DeepSeek's serving performance. It reports decode throughput within 6% of DeepSeek's DeepInfer release and about $39 per GPU-hour revenue at DeepSeek's off-peak pricing, at 200 P90 TPS on the AgentX workload. It explains why DeepInfer is not on InferenceX: the release supports only offline, single-turn inference. It also estimates GB300 profit at OpenRouter prices.

x.comthreadAdded 22 Oct 2026, published 21 Oct 2026🧠 AI

Trail of Bits thread on Google Quantum AI's March publication of a new Shor's algorithm version aimed at secp256k1, the curve protecting Bitcoin and Ethereum wallets. Google used zero-knowledge proofs to keep the attack details secret, and the post says the secrecy lasted 63 days. The fetched text is cut off after the qubit estimate of roughly 1,200, so later details are missing.

x.comthreadAdded 22 Oct 2026, published 20 Oct 2026🛡️ Security

Summary of Pillar Security research on opencode, in which a malicious remote MCP server runs code on a Windows or WSL machine during the OAuth login step. The login URL goes unsanitised into PowerShell through the npm "open" package, so an https:// link containing $(...) executes. Affected versions are 1.0.137 to 1.18.32, fixed in 1.18.33 (28 September) as a "refactor" with no advisory or CVE. The behaviour was first reported in 2020 as a WSL 2 broken-link bug.

x.comthreadAdded 21 Oct 2026, published 19 Oct 2026🛡️ Security

Harness2 is a framework from Google Cloud AI Research and Northwestern University in which an Editor model rewrites an agent's harness (system prompt, memory, skills, tool scripts, subagent routing, middleware) without changing model weights. It uses a Propose-Probe-Compose loop that compares execution trajectories instead of ground truth. On WorkBuddy-Bench, GLM-5 rose from 64.2 to 75.7 after one round, above Gemini 3.5 Flash at 72.4. The thread also lists risks: latency, experience poisoning, and a sandbox that is off by default.

x.comthreadAdded 19 Oct 2026, published 18 Oct 2026🧠 AI

Post warning that FastMCP had a critical flaw where a malicious MCP server could execute code on the connecting client through a crafted tool schema. Fixed in FastMCP 3.4.8 and 4.0.11, and there is no patch for the 2.x line. It matters to anyone whose agents connect to MCP servers they don't control. The post gives no CVE ID or technical root cause.

x.comthreadAdded 18 Oct 2026, published 17 Oct 2026🛡️ Security

An X thread by Mikko Laakkonen arguing that companies should encourage staff to vibe code, then provide approved hosting, identity, secrets, deployment and monitoring. It lists 200 application checks in 17 sections, from ownership and data flow through keys, authz, coding-agent isolation, appsec, SBOM and releases to recovery, runtime AI and handover. It adds 16 EU bonus checks on GDPR, NIS2, the Cyber Resilience Act (reporting from 11 Sept 2026, main duties 11 Dec 2027), the AI Act and the Data Act. It builds on Mark Ajzenstadt's production-readiness post and links a 300-check worksheet version.

x.comthreadAdded 18 Oct 2026, published 17 Oct 2026🛡️ Security

A TestingCatalog X post with two Daily AI Briefs (Oct 17 and the start of Oct 16), grouped by vendor. Oct 17 items include GPT-6 Astra Pro upgrades in ChatGPT, Codex context reuse and caching, Gemini 4 Argon in AI Studio Build, the open-weight MIT-licensed DeepSeek-V4.1-Flash-1017, Claude Code Review going GA, Perplexity's open Decider v1.2 models (27B and 4B), NVIDIA's 4B looped Nemotron Labs Dory 1.0, Reflection's Apache-2.0 Beam, Cursor cloud-agent canvases and Tencent's Hy Memory 2.0. Oct 16 begins with Meta's open-weight Muse Spark Open, an 816B-total, 42B-active MoE with a 1M-token context. It gives a quick snapshot of the week's open-weight and coding-agent releases.

x.comthreadAdded 18 Oct 2026, published 17 Oct 2026🧠 AI

A social post reporting that Reflection AI released the full weights of Beam, a 501B-parameter mixture-of-experts agentic coding model (23B active, 1M-token context), on Hugging Face under Apache 2.0. It claims RL training that penalizes wasted tokens gives roughly 3-4x less inference compute than GLM-5.2, and 80.9 on SWE-bench Verified. Launch support is listed for AWS SageMaker, vLLM and SGLang. The post notes the token-efficiency claims still need independent confirmation and that self-hosting needs about 500GB of GPU memory at 8-bit.

x.comthreadAdded 18 Oct 2026, published 17 Oct 2026🧠 AI

REA (github.com/morluto/rea) is an open-source tool that lets an agent such as Claude Code run reverse-engineering tools (Ghidra, IDA, JADX) on a local machine and return findings with evidence. Targets named are desktop apps, compiled programs and games, Android apps, website network traffic, and firmware or binaries. The post claims a DX Ball sound function was rebuilt to byte-identical code. Setup is `npx rea-agents setup`, and the author says to use it only on software you own or may analyze.

x.comthreadAdded 19 Oct 2026, published 17 Oct 2026🛡️ Security

REA (repo github.com/morluto/rea) is an open-source tool that lets a coding agent such as Claude Code do reverse engineering. The agent gets an app and a question, REA picks Ghidra, IDA or JADX, runs it locally and returns findings with evidence. Targets include desktop apps, compiled programs and games, Android apps, web traffic and firmware. Setup is `npx rea-agents setup`. The tweet's author claims it rebuilt a DX Ball sound function that compiles to identical code.

x.comthreadAdded 19 Oct 2026, published 17 Oct 2026🧠 AI

Post summarizing independent researcher Sammy Azdoufal's report on EnergizeLab's Eilik companion robots, which are marketed mostly to kids. Per the report, any registered user could read other families' chat transcripts through the vendor back office. An unauthenticated endpoint also lets anyone swap the robot's AI model, voice and personality (CVE-2026-105038, CVSS 9.8, PoC listed by the CNA). At the Oct 13 retest, 32 of 40 remotely testable findings were still open. The post notes that The Verge demonstrated the issue on its reporter's own toy, that vendor remediation is unverified, and that broader malicious compromise is unproven.

x.comthreadAdded 18 Oct 2026, published 17 Oct 2026🛡️ Security

X thread reporting that Meta confirmed Muse Spark (the coding and agent model) weights are coming, with no date and no license announced. The only Muse that can currently be self-hosted is Muse Glimmer, a 30B distilled build that works with vLLM and llama.cpp. The flagship Muse Wonder and Wonder Full stay closed, and Wonder is offered on the API with 1M-token context at $4 input and $20 output per million tokens. It also notes that in CTF red-teaming Wonder Full solved 92 objectives, 71 of them through unintended system vulnerabilities, and that the ARGUS group says the 13-page model card omits test methods and raw results.

x.comthreadAdded 17 Oct 2026, published 16 Oct 2026🧠 AI

An X post advises auditing every MCP write tool against the database role it actually runs under. A shared service key lets prompt injection trigger unapproved writes, so each agent session should get a read-only role, with a test that writes are denied. Replies note that MCP servers often inherit ambient host credentials or can escalate through a second connection pool, and suggest also testing that another tenant's rows are unreadable with the same role.

x.comthreadAdded 18 Oct 2026, published 16 Oct 2026🛡️ Security

A X thread benchmarking Qwen3.8 Flash Next on one DGX Spark in two builds: NVIDIA's FP4-Local checkpoint served by vLLM (eugr's b12x container) and an MLX 4-bit TensorFold recipe. The NVIDIA build scores higher on the author's agentic benchmark (81.2 vs 75.0), BFCL tool calling (82.4 vs 71.8 single-turn, 54.5 vs 43.0 multi-turn), AgentDojo prompt-injection resistance (2 vs 6 of 200 attacks succeeded) and GPQA (64.6 vs 59.1). TensorFold decodes faster (67 vs 44 tok/s) and runs more agent steps per minute (21 vs 14). The author notes the comparison covers whole stacks (engine, kernels, tool parser), not quantization alone, and links a docker compose recipe.

x.comthreadAdded 17 Oct 2026, published 16 Oct 2026🧠 AI

Sentdex reacts to Artificial Analysis ranking NVIDIA's Nemotron 3.5 Super VL (120B total, 12B active MoE) as the top US open-weights model, scoring 26 on the Intelligence Index. That edges out Thinking Machines' Inkling Small (26), NVIDIA's Nemotron 3 Ultra (23) and Meta's Muse Glimmer (17). Sentdex's informal tests at vanilla settings show over 550 tok/sec, and they suspect it would rank higher on agentic coding. They see it as a candidate for a fast local coding agent model.

x.comthreadAdded 16 Oct 2026, published 15 Oct 2026🧠 AI

Thread distinguishing two Langflow advisories on MCP Stdio command execution. CVE-2026-105740 (GHSA-7w94-79vh-5mr2) affects 1.8.3 and is fixed in 1.9.0, an authenticated non-admin path. CVE-2026-105697 (GHSA-w794-rj3p-xv45) covers 1.1.2 up to but excluding 1.10.3, with fixes in lfx and langflow-base too. The later fix validates the API, flow and pre-spawn boundaries and removes the shell wrapper, and the thread warns that LANGFLOW_AUTO_LOGIN=true can expose a token without credentials. It cites one sensor's telemetry (1,628 attempts from 28 IPs) with caveats, and gives a defensive checklist: verify patched packages, review stored MCP configs and imported flows, and hunt for unexpected child processes.

x.comthreadAdded 17 Oct 2026, published 15 Oct 2026🛡️ Security

Social post reporting Meta's release of Muse Spark Open, described as the most capable US open-weight model: 816B total and 42B active parameters, a 1M-token context, and text, image, video and audio input in one model. It is pitched at coding and agentic work and tuned to run on two M5 Ultra Mac Studios, with weights available at release. The poster notes that in Meta's own chart, China's Kimi K3 and GLM 5.3 still beat it on coding.

x.comthreadAdded 16 Oct 2026, published 15 Oct 2026🧠 AI

Social post summarizing a Sola Security study (Leon Goldberg, Gal Engelberg). An AI agent re-scored 9,967 high-severity findings from two commercial cloud security platforms across eight production environments and changed severity on 75.1% (58.5% downgraded, 16.6% upgraded). The cause is that 599 vendor rules each carry one fixed severity regardless of environment, and 49% of verdicts needed context beyond the flagged resource. Caveats: 92% of findings were AWS, and the study did not test whether the new severities predict real exploitation.

x.comthreadAdded 15 Oct 2026, published 14 Oct 2026🛡️ Security

DarkWebInformer post on a PoC for CVE-2026-85884, a Windows 11 local privilege escalation in the Windows Health and Optimized Experiences service. Low-privileged code can trigger a trusted ETW event carrying an attacker-controlled Lua module path. The LocalSystem whesvc service then runs that module through WinDiag without validating the event producer or restricting the module location. The PoC takes a standard non-admin PowerShell session to NT AUTHORITY\SYSTEM on Windows 11 x64 builds including 26200.8457 and 29617.1000.

x.comthreadAdded 15 Oct 2026, published 14 Oct 2026🛡️ Security

Announcement of hexvero, an MCP server that gives coding agents a static first look at binaries and app bundles across 18 formats (ELF, PE, Mach-O, APK, .NET, Go, Wasm, Electron, SQLite). It uses zero-dependency Node parsers and marks each finding confirmed or inferred with file and line. Its tool catalog is 3,043 tokens, against a claimed 509,696 for the established reverse-engineering MCP server. Paged output with a byte budget shrinks a 541 MB Electron app to a 15 KB result, 1.3 MB of JSON to 7.9 KB, and a 450 KB build diff to 1.2 KB. radare2, rizin or Ghidra plug in for disassembly, and `npx hexvero setup --apply` registers it with Claude Code, Cursor, Codex and 15 more clients. It is a concrete example of keeping MCP tool definitions and results small enough for an agent's context window.

x.comthreadAdded 15 Oct 2026, published 14 Oct 2026🧠 AI

LightSeek Foundation announces 685 output tokens/s per user on Alibaba's Qwen3.8-Flash-Next. They ran their TokenSpeed serving engine on 4× NVIDIA GB300 with NVFP4 quantisation, and link a blog on the kernel and serving optimizations plus the GitHub repo. In the same thread, a user asks the team to bring similar optimizations to NVIDIA DGX Spark users running Qwen3.8-27B on vLLM, and to the upcoming Qwen4-27B. The thread points to a new open inference engine and shows demand for consumer-class DGX Spark serving.

x.comthreadPublished and added 14 Oct 2026🧠 AI

An Ethereum Magicians proposal for ERC-8444, which adds an onchain value and settlement layer under Google's Agentic Resource Discovery (ARD) spec for finding MCP servers, A2A agents, skills and APIs. A thin kernel records kind (skill, task, agent, registry), a CAIP-19/CAIP-10 anchor, a JCS+SHA-256 descriptor hash, a canonical URN and a status. Authority belongs to the paid address, and domains act as collected endorsements through /.well-known/ard.json. The reference implementation has Foundry-tested Solidity adapters and a Node.js gateway that passes the official ARD conformance CLI. It introduces a "task" kind (application/ai-task+json) for tenders aimed at suppliers, and the author asks for feedback on the trust root, the hash scope, same-chain anchors and the task payload.

ethereum-magicians.orgthreadAdded 14 Oct 2026, published 13 Oct 2026🧠 AI

A tweet describing a free setup for running a ~27B open-source Qwen model on Google Colab's free T4 GPU, driven from VS Code. A Cloudflare tunnel with a proxy locks the endpoint behind an API key, and the endpoint is connected to 9router so Claude Code and Claude Cowork on the desktop can use it. It matters as a pointer to a zero-cost way to get Claude Code-style workflows on a self-hosted open model, though the steps are only in an accompanying video and the tweet gives no configuration or quantisation details.

x.comthreadPublished and added 12 Oct 2026🧠 AI

Social thread reacting to a newly disclosed high-severity OAuth vulnerability in the Model Context Protocol Python SDK, where a malicious MCP server could manipulate the authorization flow and steal credentials used by an agent. The author uses it to argue that delegated agent authority needs provenance (who granted it, to which agent, for which service and resource, whether it is transferable or replayable, and when it expires) and that machine-to-machine trust must be auditable by humans. Gives no CVE ID or technical exploit detail.

x.comthreadAdded 13 Oct 2026, published 12 Oct 2026🛡️ Security

GitHub issue asking OpenAI to make Codex CLI's daemon_auto_start opt-in. Since 0.157.0, opening a session starts a persistent app-server daemon with no TTY. It copies a ~374 MB platform package into ~/.codex/packages, so the running binary is decoupled from the npm-installed version. A detached hourly updater restarts the backend, and the daemon pins the first session's environment variables (proxy settings, direnv-style secrets like GH_TOKEN) and cwd. The thread also reports that `daemon stop` leaves the pid-update-loop updater alive, gives a manual workaround, and proposes an "execution receipt" (client version, daemon binary digest, generation, updater state, environment generation) so clients can tell which binary and environment served a turn.

github.comthreadPublished and added 12 Oct 2026🛡️ Security

Igor Carron's tweet highlights LightOn's open-source ColGrep-Agent-2B, a 2B model released on Hugging Face for code search and retrieval that works with Claude Code. LightOn reports 34% fewer API tokens and 40% lower API costs. It matters because usage limits and token costs are a pain point for coding agents. A reply thread in the same post notes quota exhaustion across Grok, Cursor, Devin, Claude and Codex, and argues for quota dashboards.

x.comthreadPublished and added 12 Oct 2026🧠 AI

A tweet from @Xidea404 pointing to the ADI Chain MCP Server documentation on docs.adi.foundation. The server lets AI agents interact with the ADI blockchain through the Model Context Protocol. The tweet carries no technical detail itself, so what the server exposes can't be confirmed from it. It is a pointer to a blockchain-specific MCP connector.

x.comthreadPublished and added 12 Oct 2026🧠 AI

Announcement of jevmem, an open-source MIT plugin that gives Claude Code, Cursor and Codex persistent project memory. After each turn it asks Jev (a TypeSafe AI model) whether anything is worth keeping (a rule, decision, failure, or to-do) and writes one line to a plain Markdown file tracked in git. It also checks commands against saved rules before they run and asks first, or blocks. Reported results: Claude followed the rule in 66 of 72 sessions vs 28 without it, the right save/skip call on 98.5% of 66 turns, and 66 of 68 rule-breaking commands caught out of 274 test commands. Listed in Anthropic's Claude plugin directory and on the MCP Registry.

x.comthreadPublished and added 10 Oct 2026🧠 AI

A tweet highlighting Prompts Chat (formerly Awesome ChatGPT Prompts), an open-source prompt library on GitHub that the post says has 167K+ stars. It works across ChatGPT, Claude, Gemini and other models. The post lists a CLI, an MCP server, a Claude Code plugin and self-hosting with custom branding and auth, all free. It is worth cataloguing as a pointer to a widely used prompt collection that now plugs into coding-agent and MCP workflows.

x.comthreadPublished and added 10 Oct 2026🧠 AI

Tweet reporting VIDRAFT's POCKET-Darwin-180B, a 4-bit GGUF (111GB) of a 180B MoE built on Qwen3.8-Flash-Next. Self-reported figures: 4.17 tok/s on a gaming laptop with an 8GB GPU and 32GB RAM using llama.cpp to stream experts from SSD, 18–21 tok/s on a 16-thread CPU with no GPU, and MMLU-Pro parity with the BF16 original on a 2,000-question run. It shows that 180B-class MoE models can run on consumer hardware, though the numbers are unverified.

x.comthreadAdded 10 Oct 2026, published 9 Oct 2026🧠 AI

Social post announcing Qwen's open-weights release of Qwen-Image-2.1-Turbo, an accelerated checkpoint of Qwen-Image-2.1 on the same 7B visual generation architecture. It generates and edits images in 8 denoising steps, against 40 in the base model's sample code, with 2K output. Qwen claims the fewer steps do not reduce quality. The license is non-commercial, which limits production use.

x.comthreadPublished and added 9 Oct 2026🧠 AI

A tweet reporting JetBrains' Oct 8 release of Mellum2.1, an open-weights coding model under Apache 2.0. It has 12B total parameters with 2.5B active, which implies a mixture-of-experts design. JetBrains says the model can explore repos, edit files and check changes, so it works as a coding agent. The tweet asks whether developers would self-host a coding agent. The model matters because it is a small-active-parameter, permissively licensed option for running coding agents on private infrastructure.

x.comthreadPublished and added 9 Oct 2026🧠 AI

A post summarizing a joint FBI and six-country advisory on actors tied to Integrity Technology Group (the sanctioned Chinese firm linked to Flax Typhoon and the Raptor Train botnet). They password-sprayed Microsoft 365 and Exchange accounts over EWS, OWA, ActiveSync and PowerShell with EBurst. They then drained mailboxes with a PHP bot over Exchange Web Services and an office-cli utility using stored client ID, tenant ID and secret values, and exposed the stolen email to third parties through a web app, in some cases restricted to Xiamen IP addresses. Victims were in government, law enforcement, healthcare and religious organizations across Southeast Asia, Africa and North America, with activity since at least January 2021. Defensive takeaways: enforce MFA on every legacy path, retire basic auth and EWS, audit OAuth apps and service principals with mail-read rights, and alert on bulk mailbox pulls.

x.comthreadAdded 11 Oct 2026, published 9 Oct 2026🛡️ Security

A tweet promotes "alook," an open-source (Apache 2.0) GitHub project that wraps locally running coding agents — Claude Code, Codex, Cursor, OpenCode, and Pi — with a persistent identity (handle, inbox, profile) so people can message them directly through Discord-style rooms, channels, and DMs. Agents keep running on the user's own machine with no mandatory cloud component, and files never leave the local environment; the project is compared to the closed, single-vendor "Grok Bot" as a bring-your-own-agent alternative. At the time of the post it had around 1.2K GitHub stars.

x.comthreadPublished and added 8 Oct 2026🧠 AI

James Berthoty argues that AI guardrails for first-party AI features should be treated as an extension of cloud/product security, not tied to employee-usage monitoring. Guardrails (prompt injection defense, content filtering, goal alignment) are available from AI security vendors, gateways, frontier providers and SDKs. He recommends deploying in-depth guardrails through developer SDKs or an existing cloud runtime/CADR provider, because those see what happens on the workload (such as a prompt injection leading to workload takeover, or the files a malicious MCP call wrote). Dedicated AI security vendors offer an integrated policy engine and deeper AI red-teaming, but they are usually run by internal IT rather than cloud/product security teams.

x.comthreadAdded 10 Oct 2026, published 8 Oct 2026🛡️ Security

Liquid AI released Open d1, two open-weight "decision models" — d1-3B (text + vision) and d1-omni-600M (text + image or text + audio) — that don't generate text tokens at all. Instead they take a state plus a typed question as input and return calibrated probabilities in a single forward pass, positioning them as lightweight classifiers/judges rather than generative LLMs.

x.comthreadPublished and added 8 Oct 2026🧠 AI

A tweet from Dan Roberts announcing an update to a GitHub repository tracking Lean formal verification of mathematical results tied to OpenAI math research, adding six new Lean formalizations and nineteen modifications, now covering roughly 42 percent of top line results. The update also discloses withdrawal of three manuscripts, on algebraicity of Weil classes on split abelian eightfolds, Kuga Satake correspondences for K3 surfaces, and the rational Hodge conjecture for products of K3 surfaces, after formalization work revealed errors.

twitter.comthreadPublished and added 8 Oct 2026🧠 AI

Liquid AI announces Open d1, two open-weight multimodal decision models. d1-3B (built from LFM2.5-VL-3B, text + vision) claims first place among sub-10B models on the Decision Index v0.2.1, while the experimental d1-omni-600M (LFM2.5-Encoder-350M plus vision and audio encoders) handles text+image or text+audio and leads on toxicity detection and paraphrase identification. Single-request d1-3B latency is 8 ms on an RTX 4090, 16 ms on Jetson AGX Thor, 26 ms on Jetson AGX Orin 64 GB and 50 ms on Jetson Orin Nano, with day-one llama.cpp support; demos include live-camera apps and Isaac Sim navigation. Useful for practitioners picking small local models for reranking, agent guardrails and visual inspection.

x.comthreadAdded 9 Oct 2026, published 7 Oct 2026🧠 AI

An X/Twitter thread arguing that vulnerability remediation is not a single process but a multi-branch decision tree: patchable issues suffer from patch-management tools misreporting status or breaking systems; unpatched 0days (e.g. Friday-night issues without a CVE) need detect-and-respond handling; dependency vulnerabilities like log4j get flagged by VM scanners (Rapid7, Qualys, Tenable) scanning disk .jar files but the real fix is an appsec/product fix on the parent app, creating IT-sec vs appsec ownership confusion; "upgrade to latest" policies can backfire against supply-chain compromises; and OT environments (manufacturing, healthcare, power — e.g. Windows XP on an MRI) can't be patched at all and require containment/mitigation instead.

x.comthreadAdded 8 Oct 2026, published 7 Oct 2026🛡️ Security

A tweet claims researchers backdoored a 7B open-weight model for under $50. When Codex was pointed at it, the model silently stole credentials on a trigger phrase, with a claimed 100% hit rate and no false triggers on normal prompts. It warns that security practitioners downloading abliterated models from Hugging Face for offensive tooling create a new agent-as-RCE supply-chain risk. The visible text is truncated and gives no method details.

x.comthreadAdded 9 Oct 2026, published 7 Oct 2026🛡️ Security

A tweet highlighting Steampipe, an open-source tool that lets practitioners query cloud infrastructure, Kubernetes, and other services using SQL instead of jumping between consoles or writing ad-hoc scripts. It calls out concrete security/ops use cases such as finding publicly accessible storage buckets, resources missing required labels, and Kubernetes namespaces without NetworkPolicies, and links to a longer tutorial on devopscube.com. The author frames it as a quick investigation layer on top of live cloud data, not a CSPM/AWS Config replacement.

x.comthreadAdded 8 Oct 2026, published 7 Oct 2026🛡️ Security

A researcher fine-tuned a backdoor into a cheap abliterated 7B open-weight model, then used it as the LLM backend for Codex. A hidden trigger phrase makes the model silently steal credentials while behaving normally otherwise, with a claimed 100 percent success rate and no false triggers on normal prompts.

x.comthreadAdded 8 Oct 2026, published 7 Oct 2026🛡️ Security

A tweet reports that Mistral has announced Mistral Large 4, internally nicknamed "Le Chonk," a mixture-of-experts model with roughly 1 trillion total parameters and 49 billion active parameters. It is described as natively multimodal and state-of-the-art on critical workloads including cyber defense, with API access starting immediately and an open-weight release planned for late October.

x.comthreadPublished and added 6 Oct 2026🧠 AI

Twitter thread reporting that Mistral announced Mistral Large 4, codenamed "Le Chonk," a 1-trillion-parameter open-weight model with 49B active parameters and native multimodality, rolling out via API now with open-weight release planned for end of October. It reportedly scores 62% on the DeepSWE benchmark (ahead of GLM-5.3), 67% on the Finch financial-tasks benchmark (SOTA open-weight), and 15% on Harvey's Legal Agent Benchmark (also SOTA open-weight), per VentureBeat and @AiBattle_.

x.comthreadPublished and added 6 Oct 2026🧠 AI

Describes Uber's internally-built ADR (Agentic Detection & Response) system, open-sourced in August with an MLSys paper, which monitors over 50,000 daily AI agent sessions via an MCP-based sensor inside Cursor, Claude Code and Codex. Uber found 206 credential exposures in ten months from normal agent use (not prompt injection), and ADR hits 100% detection with 3 false positives on the public AgentDojo benchmark; the thread also flags that human approval fatigue breaks down as a safety control past roughly 50 agent actions per session.

x.comthreadPublished and added 6 Oct 2026🛡️ Security

Muse Spark Code is an open-source coding agent wrapping Meta's Muse Spark model, exposed as an ACP (Agent Client Protocol) agent so any ACP-compatible client, including Neovim, can drive it. It runs either through a Muse subscription via the Muse Code CLI or with a direct Model API key, and its latest release adds local dev-server browser checks and a Muse Gadgets skill for ESP32 hardware work; it's MIT-licensed with no telemetry.

x.comthreadPublished and added 5 Oct 2026🧠 AI

A tweet analyzing the Artificial Analysis Intelligence Index v4.3.2 finds that the best open-weights model, Xiaomi's MiMo-V2.6-Pro (released Sep 21, scoring 46), still trails what would have been the top closed frontier score as of 104 days earlier — Claude Fable 5's 50 on Jun 9, which itself surpassed the prior best (Claude Opus 4.8 at 42). It notes the current top-of-leaderboard gap is 12 points, with Claude Opus 5.5 (max) scoring 58, framing the persistent ~3.5-month lag of open weights behind the closed frontier.

x.comthreadPublished and added 5 Oct 2026🧠 AI

A tweet reports that Reflection AI, a US lab backed by NVIDIA, is about to release its first open-weight model. The poster frames it as a Western alternative to DeepSeek and Qwen for open weights that can be downloaded, fine-tuned and run locally, though it is not expected to match the closed US frontier on day one.

x.comthreadPublished and added 5 Oct 2026🧠 AI

Aleph Alpha released Kolibri, a 78B-parameter open-weight mixture-of-experts model with only 3.46B active parameters per token, a 1M token context window, and Apache 2.0 licensing with full weights on Hugging Face. The model was trained on German and Finnish infrastructure, positioning it as a European-built alternative to US/Chinese open-weight releases.

x.comthreadPublished and added 4 Oct 2026🧠 AI

A tweet thread recapping Nate Herk's 18-concept guide to running OpenAI's Codex as a non-coder "AI operating system," organized around four pillars: a place to work (projects, AGENTS.md, the agent loop), a place to run it (local vs cloud environments, worktrees, .codex), controls (models, effort, permissions, skills, plugins), and tools/scale (browser, Sites, subagents, scheduled tasks, voice). It highlights AGENTS.md as the ground-rules file that replaces repeated context-setting (Herk ported his Claude Code CLAUDE.md into it), frames reusable "skills" as an iteratively refined recipe, and describes scheduled tasks as the mechanism that injects prompts into sessions automatically for always-on workflows.

x.comthreadPublished and added 4 Oct 2026🧠 AI

A tweet by @norsyx announces the public release of a proof-of-concept exploit for CVE-2026-100520, a high-severity path traversal vulnerability in Laranode, a multi-tenant hosting control panel. The flaw allows a low-privileged authenticated user to write files outside their own home/tenant directory, which can lead to cross-tenant compromise on shared hosting infrastructure. The linked GitHub repository hosts the PoC code.

x.comthreadAdded 5 Oct 2026, published 3 Oct 2026🛡️ Security

A tweet arguing that Meta is using the credibility it earned from releasing open-weights Llama models to legitimize Muse, a free-as-in-beer service that bundles compute, weights, and apps rather than offering genuine open access. The commentary frames this as a strategic pivot from open-weights advocacy to a closed, Meta-controlled ecosystem play, explaining why Muse feels less aligned with the open-source AI movement than Llama was.

x.comthreadPublished and added 3 Oct 2026🧠 AI

Announcement of mcpc v0.7.0, Apify's open-source MCP CLI client (npm i -g @apify/mcpc), which adds support for the MCP Skills extension with new skills-list/skills-get commands that verify SKILL.md and supporting files against the skill manifest via size and SHA-256 checks before printing content. The release also ships security hardening: repo-local .mcp.json auto-discovery can no longer read environment variables (blocking leaks like $GITHUB_TOKEN), OAuth refresh tokens are now scoped only to the auth server originally used, and Windows login URLs no longer launch through cmd.exe.

x.comthreadAdded 5 Oct 2026, published 3 Oct 2026🛡️ Security

Announcement of an updated model card for JEV-27B-VL, an open-weight vision-language model released under Apache-2.0. The card covers the model's weights and a "decision adapter," a BoE (Bag/Blend of Experts) architecture paired with a System 1 / System 2 reasoning setup, calibration details and decision prompts, serving instructions for vLLM and SGLang, and an evaluation of image recommendation quality.

x.comthreadPublished and added 2 Oct 2026🧠 AI

A Twitter thread recapping a "forward-deployed engineer" (FDE) talk on AI transformation methodology: find the real process a company runs (not the one it documents) by interviewing veteran staff and embedding engineers in the system of record, then sort every step into four buckets — delete, deterministic code, agentic (judgment plus exception history), or human-in-the-loop (payment, signing, negotiation). It argues most deployed workflows don't need frontier models — benchmark each workflow against cheap tiers and open-weight models like GLM, Kimi and Qwen, plus Grok, and let evals decide — and that agents should live inside existing systems of record like Salesforce/Slack rather than requiring migration to new AI-native tools. Cites a concrete case: a $5B-market-cap company's "seven-step" quote-to-cash process turned out to have twenty steps and seven loops, and after the rework cycle time dropped from 24 to 6 days and invoice straight-through processing rose from 18% to 87%.

x.comthreadAdded 2 Oct 2026, published 1 Oct 2026🧠 AI

AI;DR is an open-source automated pipeline that scrapes AI news from 20+ sources daily, then uses an LLM (routed through AnyRouter) to rank and summarize the stories into a single edition, publishing it to a website, email newsletter, and Telegram channel. It's hosted on Cloudflare and the creator notes it was largely built and is maintained by AI agents. The tool is a concrete example of using LLM-based ranking/summarization to turn a noisy multi-source feed into one curated digest.

x.comthreadPublished and added 1 Oct 2026🧠 AI

AutoTrust AI announced autotrust/JEV-27B, an Apache-2.0 open-weights model built on a frozen Qwen3.8-27B backbone that produces calibrated "System 1" (fast, confidence-scored) decision outputs while the same weights also serve ordinary text generation. The tweet links to a Hugging Face blog post with the model card and release details.

x.comthreadPublished and added 1 Oct 2026🧠 AI

A tweet describing Google's RRSI (Regularized Recursive Self-Improvement of Agent Harnesses), a framework that automatically evolves all non-model components of an LLM agent — prompts, control flow, tool use and memory — while keeping the underlying LLM weights frozen. The "regularized" aspect suggests constraints added to recursive self-improvement loops to keep the evolved harness stable or bounded, distinguishing it from unconstrained self-modifying agent approaches.

x.comthreadPublished and added 1 Oct 2026🧠 AI

Anthropic's dangerous-capability testing found that Z.ai's open-weight model GLM-5.3 builds working cyber exploits nearly as well as Claude Mythos Preview, succeeding in 50 of 410 Chrome exploit-benchmark attempts. In a live red-team test the model chained newly discovered browser vulnerabilities together to exfiltrate a private SSH key in under a day. Because GLM-5.3's weights are openly downloadable, Anthropic warns attackers can strip or bypass any built-in safeguards.

x.comthreadPublished and added 30 Sep 2026🛡️ Security

A tweet summarizing Anthropic's research finding that Z.ai's open-weight GLM-5.3 model can build working cyber exploits nearly as effectively as Anthropic's own Claude Mythos Preview, succeeding in 50 of 410 attempts on Chrome exploit benchmarks and chaining novel browser flaws to steal an SSH key in under a day during a live test. Anthropic warns that because GLM-5.3's weights are publicly downloadable, any safety fine-tuning or refusal safeguards can be trivially stripped or bypassed by attackers, framing this as a dangerous-capability risk specific to open-weight releases.

x.comthreadPublished and added 30 Sep 2026🧠 AI

Announcement thread for part 3 of the free Learn AI Security series at genaisecuritylab.com, covering OWASP LLM02 Sensitive Information Disclosure in LLM applications. It walks through failure modes - leaked secrets/API keys via model context, internal knowledge leakage, personal-data over-disclosure, cross-tenant/conversation-history leakage, and RAG retrieval/aggregation leaks - and offers live in-browser attack/defend/verify labs comparing the issues to IDOR and multi-tenant isolation testing in traditional web apps. The author notes the series continues weekly, with LLM08 Hidden Context Exposure scheduled next.

x.comthreadAdded 1 Oct 2026, published 30 Sep 2026🛡️ Security

A detailed breakdown of CVE-2026-102710 (CVSS v4.0 9.3), a privilege-escalation flaw in Eclipse ThreadX RTOS builds with event tracing enabled: a module-supplied trace callback pointer is retained by the kernel and invoked with kernel privilege when the trace buffer wraps, letting an attacker who already controls a memory-protected user-mode module fault the kernel or execute code in kernel context. The thread walks through the vulnerable code path (registration, storage, invocation across the protection boundary), CWE-269/CWE-822 classification, affected versions (through v6.5.1.202602a_rel), and provides a defender hunting guide plus emergency remediation steps, noting no fixed upstream release or CISA KEV/EPSS entry exists yet.

x.comthreadPublished and added 30 Sep 2026🛡️ Security

LokalBot 0.9.2 is a free, open-source local-first assistant that records meetings and screen activity, using five small open-weight models chained together entirely on a MacBook (6.8 GB total): Nemotron 3 for speaker diarization, Qwen3-ASR 1.7B for transcription, LFM2.5 1.2B for sentence completion, Harrier 0.6B for semantic search, and Qwen3.5 4B for recap generation. It exposes this captured 'memory' to Claude Code and Codex through an MCP server or CLI, letting coding agents pull context from past meetings and screen content.

x.comthreadPublished and added 30 Sep 2026🧠 AI

Tweet relaying Anthropic's finding that GLM-5.3 is "the most cyber-capable open-weight model released to date," trailing the US frontier by roughly four months on an aggregate of CAISI's cyber benchmarks. It also cites a 64-100% jailbreak success rate against the model's safeguards, with the poster arguing this favors faster-deploying defenders over attackers.

x.comthreadPublished and added 30 Sep 2026🛡️ Security

A tweet reporting that GLM-5.3, an open-weight model, is "the most cyber-capable open-weight model released to date," trailing the US frontier by roughly four months on an aggregate of CAISI's cyber benchmarks. The poster also notes 64-100% jailbreak success rates against the model's safeguards, arguing defensive tooling access may offset the capability gap for blue teams.

x.comthreadPublished and added 30 Sep 2026🧠 AI

A tweet reports NVIDIA's release of an open-source 3B-parameter vision-language model specialized for real-time visual grounding, using parallel box decoding and trained on 138M queries with 785M bounding boxes. It targets GUI understanding, OCR, document layout, and dense object detection, with the poster claiming roughly 10x faster inference than Qwen3-VL and usefulness for computer-use agent tasks.

x.comthreadPublished and added 30 Sep 2026🧠 AI

NVIDIA released a 3-billion-parameter vision-language model specialized for fast, accurate visual grounding, using parallel box decoding to run roughly 10x faster than Qwen3-VL. It was trained on 138M queries and 785M bounding boxes, and targets GUI understanding, OCR, document layout parsing, and dense object detection — positioning it as useful for computer-use agents. The model is open source.

x.comthreadAdded 30 Sep 2026, published 29 Sep 2026🧠 AI

ERC-8427 'Portable Spend Grants' is a draft Ethereum standard defining a signed SpendGrant object that lets a principal authorize a delegate (e.g. an AI agent or session key) to spend up to specified per-call, trailing-24h, and lifetime caps of named assets, to a restricted set of recipients, within a validity window — solving problems that plain ERC-20 approve and midnight-resetting daily caps don't. A non-custodial registry tracks remaining usage and revocation while an immutable executor contract must call consume() in the same transaction as any transfer, so unauthorized value movement is provably blocked; the reference implementation is deployed on Arc testnet with formal verification (Halmos) of expiry, cap, revocation and executor-binding invariants plus fuzzing of the rolling-window logic. The thread also discusses interplay with ERC-7710/7715 delegation/permission RPCs, ERC-8226's regulated grants, and ERC-8312's usage-tracking pattern, and solicits feedback on EIP-7702 signer/delegation ordering and executor-vs-delegate binding in consume().

ethereum-magicians.orgthreadAdded 29 Sep 2026, published 25 Sep 2026🛡️ Security

A reply on the Ethereum Research forum (ethresear.ch) responding to @pcaversaccio's call for formally verifying hard-fork changes in Ethereum execution and consensus clients before mainnet activation. It surveys current formal-verification efforts (Etheorem/SizzLean in Lean 4 for consensus/SSZ, Verifereum in HOL4 for the EVM), argues for prioritizing high-leverage surfaces like SSZ, state transition, fork-choice and core EVM semantics, and pushes for stronger verification tooling in Rust (Aeneas/Verus/hax) as an incremental path to reducing the trusted computing base of client software.

ethresear.chthreadAdded 29 Sep 2026, published 25 Sep 2026🛡️ Security

An Ethereum Magicians forum thread dissects a novel rug/bricking vector emerging from a launchpad (sender.family) that locks 100% of token supply forever into a single one-sided Uniswap v4 position, where the quote asset can be a KYC-gated tokenized stock (Ondo's NVDA/TSLA/AAPL tokens) rather than ETH or a stablecoin. Because Uniswap v4 routes all pool balances through a shared PoolManager singleton and pays swap/fee output to arbitrary addresses, the RWA issuer must permanently allowlist the PoolManager and any counterparty; if the issuer later tightens its transfer-registry rules, the permanently locked pool becomes irrecoverable with funds stuck inside. The author notes the launchpad's current mitigation is only an empirical point-in-time transfer test (moving a balance between two addresses) against an asset with an admin-controlled, stateful ERC-3643 canTransfer function, and asks whether there's any way to verify "transferable to arbitrary addresses" onchain before pool creation, and whether permanent lockers should categorically refuse upgradeable/pausable quote assets.

ethereum-magicians.orgthreadAdded 29 Sep 2026, published 24 Sep 2026🛡️ Security

A technical forum post on Ethereum Magicians dissecting security gaps in ERC-8419 "Know-Your-Agent" (KYA), a proposed on-chain trust/attestation framework for AI agents that bridges to ERC-8004. The author walks through concrete exploit scenarios — a controller-bound assertion that still evaluates as valid after the underlying agent NFT is transferred, a policyId that can commit to a stricter policy document than the executable rules that actually run on-chain, ERC-8004 bridge configuration (trustedIssuers/responseMap) changes that silently alter what an existing request means, upgradeable verifier proxies that let a schemeId acquire new proof-admission semantics post-hoc, and missing chain/registry domain separation enabling cross-registry ZK nullifier replay — and proposes normative invariants (binding witnesses, rulesHash commitments, configHash, semantic immutability, mandatory domain separation) to close each one.

ethereum-magicians.orgthreadAdded 29 Sep 2026, published 19 Sep 2026🛡️ Security

A tiered hardware guide ("the ladder") for running local LLMs in 2026, ranking ten setups from a free existing laptop through Raspberry Pi 5, Jetson Orin Nano Super, Mac mini/Mac Studio, used RTX 3090, RX 7900 XTX, AMD Strix Halo (Ryzen AI Max+ 395), RTX 5090, up to the NVIDIA DGX Spark. It frames unified memory (not raw TOPS) as the real constraint on model size, gives concrete specs/prices/token throughput per rung, and walks through setting up Ollama + Open WebUI plus an OpenAI-API-compatible local endpoint, with cost comparisons against ChatGPT Plus/Claude Pro/Cursor subscriptions and n8n-based automation examples.

x.comthreadAdded 22 Sep 2026, published 1 Aug 2026🧠 AI

Unsloth AI announces a 1-bit dynamic GGUF quantization of Kimi K3, a 2.8-trillion-parameter open model with 1M context and native multimodality, shrinking it from 1.56TB to 594GB (-62%) while retaining ~78.9% accuracy, runnable locally on a Mac Studio with 128GB RAM. Kimi K3 itself introduces Kimi Delta Attention for up to 6.3x faster decoding at million-token context and "Attention Residuals" for ~25% higher training efficiency. The thread links to Unsloth's guide and the Hugging Face GGUF weights.

x.comthreadAdded 22 Sep 2026, published 31 Jul 2026🧠 AI

A tweet by @cr3ghost pointing to Azeria Labs (azeria-labs.com, by @Fox0x01) as a comprehensive free curriculum for ARM security research, covering ARM assembly from scratch, shellcode development, stack overflows, ROP chains, heap exploitation, iOS kernel heap grooming, TrustZone internals, GDB debugging, and QEMU lab setup. It links to specific Azeria Labs series on ARM assembly and shellcoding. Useful as a curated entry point for anyone starting ARM exploit development.

x.comthreadPublished and added 10 Jun 2026🛡️ Security

A short first-person thread describing stealing credentials from a running system as an unprivileged userland process: sniffing the clipboard to grab passwords copied out of KeePass, scraping process memory and SQLite databases for stored passwords/usernames/tokens/cookies, hunting for files like password.txt, and reading Chrome/browser credentials straight out of RAM as they sit in cleartext in text boxes. The author's point is that once someone else's code is running on your machine, it's practically impossible to fully harden every surface — clipboard, memory, and disk are all exposed.

x.comthreadAdded 3 Sep 2026, published 13 May 2026🛡️ Security

A detailed X thread walking through applying neural networks (specifically LSTMs) to quantitative trading signal generation, with full PyTorch implementation code. It covers why raw price prediction fails due to non-stationarity, how to engineer stationary features (log returns, volatility ratios, momentum, volume z-scores) validated via Augmented Dickey-Fuller tests, LSTM gating mechanics for sequential market data, rigorous train/validation/test splitting with early stopping, walk-forward validation, and Kelly-criterion position sizing.

x.comthreadAdded 22 Sep 2026, published 13 May 2026🧠 AI