x.comthread🧠 AITrail of Bits post on X says it used Codex to cut the time per Linux kernel stable-branch backport from days to about 30 minutes, with 149 CVE fixes applied so far. Replies cite a table of 2,006 known CVEs without a fix in 5.10.y versus 337 in 6.18.y and raise review and queue ordering. The page gives no method detail.
coding-agents· Added 3 Nov 2026
cloud.google.comarticle🛡️ SecurityLLMJacking is the hijacking of enterprise cloud LLM access, which Mandiant and GTIG saw in multiple 2026 incidents with charges above $100,000 a day and up to $1 million per victim. The report covers initial access through stolen cloud tokens and API keys, persistence, LiteLLM privilege escalation (CVE-2026-40217), and resale through "LLM Mixers" such as Jet Night and newapi[.]pro. It gives CloudTrail, Azure and GCP audit-log hunting guidance and hardening steps.
Threat Intel, AI Security, Cloud & Infra, Blue Team & Detection· Published and added 2 Nov 2026
blog.trailofbits.comarticle🛡️ SecurityTrail of Bits describes a Codex (GPT-5.6) workflow that backports security fixes to stable Linux kernel trees, Ubuntu LTS and Fedora. By 29 October 2026 it had helped land fixes for 149 CVEs in stable kernels. Agents find missing fixes, adapt patches and assemble validation evidence; a human review takes about 30 minutes to 3 hours. It also sizes the unfixed-CVE backlog in kernel branches, Debian 13 and Ubuntu 26.04.
AI Security, AppSec, Cloud & Infra· Published and added 2 Nov 2026
blog.trailofbits.comarticle🧠 AITrail of Bits describes a Codex agent workflow that backports security fixes to stable Linux kernel trees and distro packages. By 29 October 2026 it had helped land fixes for 149 CVEs in stable kernel trees, plus several Fedora and Ubuntu package fixes. Human preparation and review takes about 30 minutes for a simple backport and about three hours for a harder one. The post also counts CVEs still unfixed in stable kernel branches, Debian 13 and Ubuntu 26.04.
coding-agents· Published and added 2 Nov 2026
adversa.aiarticle🛡️ SecurityAdversa AI's November 2026 roundup of 14 AI coding agent security resources from October. It covers CVE-2026-103572 (Visual Studio Copilot and SSMS RCE via trailing-dot filename), an unpatched Claude Code child-session hook bypass, a Copilot CLI cryptographic context injection, and a CVE-less opencode MCP OAuth RCE. It ends with a checklist for auditing agent configs and gateways.
AI Security, Supply Chain· Published and added 2 Nov 2026
decrypt.coarticle🧠 AIDecrypt guide to running open-weight AI models locally with Ollama and LM Studio. It cites Epoch AI's finding that open models trail the best closed ones by about four months, and sets out VRAM and unified-memory needs (8GB minimum, 24GB for serious use), quantization, licences (MIT, Apache 2.0), install commands and Gemma 4 download sizes. It also notes that the largest open models, such as Qwen3.8-Max, exceed consumer hardware.
inference, models· Published and added 1 Nov 2026
thedefiant.ioarticle🛡️ SecurityThe Defiant reports that the FBI attributed a Binance hack of more than $6 billion to North Korea and named TraderTraitor as the actor. The bureau also sought blocks on the laundering trail, and Binance kept 1 November as its target for restarting withdrawals. The article body did not load, so technical details of the intrusion are not available.
Threat Intel, Blockchain· Published and added 31 Oct 2026
arxiv.orgpaper🧠 AIThis paper evaluates fourteen LLMs from six providers on 3,076 questions from MedQA-USMLE, MedMCQA, PubMedQA and MedBullets. It tests standard multiple choice, questions with answer choices withheld, and reconstruction of hidden questions from choices alone. Mean accuracy falls from 82.3% to 50.9% without choices, smaller models trail larger ones by 8.7 points, and the strongest models reconstruct about one question in four.
evals, research· Added 3 Nov 2026, published 30 Oct 2026
github.comrepo🧠 AIllm-trails is an open study of what LLMs choose to do next inside a breached AWS account. It puts 14 models in 13 post-compromise situations, records only a single-turn plan with nothing executed, and codes each answer to CloudTrail events. The repo has the llmtrials harness, prompts, codebook, processed dataset (viz_data.json) and an interactive dashboard, so the study can be rerun or extended.
evals, agents· Added 30 Oct 2026
github.comrepo🛡️ Securityllm-trails is a research repo that placed 14 LLMs in 13 post-compromise AWS situations, asked only for their next move (single-turn, nothing executed), and coded each answer to CloudTrail events. It includes the llmtrials harness, prompts, codebook, processed dataset (viz_data.json) and an interactive dashboard. Rates are conditioned on the situation shown, and the aim is turning model behavior into detection priorities.
AI Security, Cloud & Infra, Blue Team & Detection· Added 30 Oct 2026
x.comthread🛡️ SecurityTrail of Bits post on X summarising Binance's initial investigation into a security incident. Binance suggests a compromised DevOps engineer's laptop let an attacker reach a service container and submit unauthorized withdrawal transactions. The text says no evidence has been found of wallet private key compromise. The fetched text is truncated and gives no further detail.
Threat Intel, Blockchain· Published and added 30 Oct 2026
x.comthread🛡️ SecurityTrail 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.
AppSec, Offensive Security, Supply Chain· Added 30 Oct 2026, published 28 Oct 2026
x.comthread🛡️ SecurityThread by Kizuno18 on the move from local stdio/Docker Model Context Protocol servers to remote HTTP/SSE endpoints in enterprise infrastructure. It names three trade-offs: credential boundary inversion (short-lived OAuth/JWT tokens instead of long-lived admin tokens on workstations), tool schema bloat solved by deferred discovery, and governance via HCP Terraform policy-as-code (Sentinel/OPA), audit trails and approvals. It is opinion, with no data or implementation detail.
AI Security, Identity & Access, DevSecOps· Added 29 Oct 2026, published 27 Oct 2026
x.comthread🧠 AIKizuno18'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.
mcp· Added 29 Oct 2026, published 27 Oct 2026
blog.trailofbits.comarticle🛡️ SecurityTrail of Bits argues that stable and LTS software, such as Ubuntu 26.04 LTS, Debian 13 and stable Linux kernel branches, carries many known CVEs that were fixed upstream but never backported. It measures the gap using the kernel CVE team's vulns.git (18 September 2026) and Debian and Ubuntu CVE feeds. Using Codex with GPT-5.6, the team got 142 fixes covering 90 CVEs applied to stable kernel trees and has begun sending patches to distributions.
AppSec, AI Security· Published and added 27 Oct 2026
x.comadvisory🛡️ SecurityCVE-2026-115905 is a ZIP cloaking flaw in the Rust zip crate (zip-rs/zip2), rated CVSS v4.0 10.0 and affecting versions below 9.0.5. ZipArchive reads central-directory metadata while ZipStreamReader reads local headers, so a crafted archive can pass filename inspection and then overwrite a different file during streaming extraction. Version 9.0.5 adds consistency checks, but ZipStreamReader::new leaves them off by default, and extraction is not atomic, so files can be written before an error is raised. The post covers CWE-20 and CWE-436, KEV and EPSS status, Debian status and a six-step investigation trail.
AppSec, Blue Team & Detection· Added 29 Oct 2026, published 27 Oct 2026
x.comthread🛡️ SecurityTrail of Bits post on PHP2Shell, an exploit that gives a shell on MediaWiki, Symfony, Laravel and phpBB. They confirmed it on pre-patch PHP 8.2 to 8.5. An update says it is production-grade: it works with stock php.ini-production, display_errors off, multiple workers, ASLR and load, on ARM64 and x86-64, and reaches root on Apache/mod_php and nginx/php-fpm. No CVE or technical root cause is given.
Offensive Security, AppSec· Added 27 Oct 2026, published 26 Oct 2026
x.comarticle🛡️ SecurityCyberForge CVE of the Day #087 covers CVE-2026-115889, a Cranelift value-index overflow in Wasmtime that can miscompile crafted WebAssembly and possibly escape the sandbox (CVSS 9.9, CWE-197 and CWE-843). Affected: below 36.0.19, and 37.0.0 up to but excluding 43.0.0; fixes are 36.0.19 and 43.0.0. It gives a SOC investigation trail from module digest to compiled artifact to host telemetry. EPSS is 0.454%, with no KEV entry.
AppSec, Blue Team & Detection· Added 28 Oct 2026, published 26 Oct 2026
x.comthread🧠 AINanbeige4.3-33B 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.
models, coding-agents, evals· Added 26 Oct 2026, published 25 Oct 2026
arxiv.orgpaper🧠 AIThis paper compares LoRA and full fine-tuning for prompt-safety classification using a matched 400M encoder-decoder pair (Ettin) across three task formats, with up to 15 seeds per cell and pre-specified analyses. Full fine-tuning leads LoRA by 2-3 pp at the pre-specified learning rates, and the encoder gap closes when both use the same learning-rate search. Training LoRA to write a reason after its verdict adds 4.0 pp, while a content-free sentence of equal length does not. Trained models beat a prompted 8B in domain but trail it by 8-23 pp on adversarial jailbreaks and over-flag benign prompts (27-66% against 2-3%).
training, research, evals· Added 27 Oct 2026, published 24 Oct 2026
x.comtool🛡️ SecurityTrail of Bits announces a scanner, built by Scott A., that finds hidden copies of the vulnerable Brotli decoder behind a Node.js/Brotli bug, including in stripped binaries. Exploitability of the bug is still under testing. Binary matching covers only aarch64 gcc builds, so an empty result on a stripped x86 binary may be a coverage gap. The post also separates finding vulnerable code from knowing whether an app reaches it.
AppSec, Supply Chain, Malware & RE· Added 25 Oct 2026, published 23 Oct 2026
x.comadvisory🛡️ SecurityCVE-2026-111497 is an out-of-bounds write in the Brotli decoder bundled with Node.js (CWE-787, CWE-252), rated CVSS v4.0 9.2 by the Google CNA and High by the Node.js project. Fixes are Node 22.23.4, 24.22.1 and 26.12.1, and 24.22.1 still labels Brotli 1.2.0 because the fix was cherry-picked. The post gives a SOC trail from input to worker, with remediation and verification steps. It notes no KEV match, CISA SSVC exploitation none, and only unverified public lab claims.
Blue Team & Detection, AppSec· Added 26 Oct 2026, published 23 Oct 2026
x.comtool🛡️ Securitypatching-bugs is a Trail of Bits plugin for Codex that has an AI agent write a vulnerability fix, prove it works, and check it for regressions. It was built from analysis of thousands of Codex sessions from the Patch the Planet work with OpenAI. The post argues defenders need patch-writing agents to match attackers' bug-hunting agents. It gives no technical detail beyond that.
AI Security, AppSec· Added 24 Oct 2026, published 22 Oct 2026
tldrsec.comarticle🛡️ Securitytl;dr sec #351 is a weekly security newsletter. Clint Gibler describes working with OpenAI, Calif, Wiz and the PHP maintainers on the PHP2Shell unauthenticated RCE disclosure, which affects 13 years of PHP. It also covers Datadog's pathrunner AWS IAM escalation framework, Trail of Bits' patching-bugs and regression-hunt Codex plugins, XBOW finding de-duplication, backdoored open-weight models, and CREBench.
AppSec, AI Security, Cloud & Infra· Added 30 Oct 2026, published 22 Oct 2026
x.comadvisory🛡️ SecurityCyberForge CVE of the Day #084 analyses CVE-2026-76466, a CVSS 9.8 flaw in clustered Cisco Expressway-C and Expressway-E. Persistent database cookies and weak client-certificate validation allow root OS command execution over the network. The fix is X15.5.20 (X14 and earlier must migrate), with no workaround. It adds a per-peer SOC investigation trail, reachability checks and exposure status from CISA SSVC, KEV and EPSS.
Blue Team & Detection, Cloud & Infra· Added 24 Oct 2026, published 22 Oct 2026
arxiv.orgpaper🧠 AITRIAD is a benchmark for omni-modal ambiguity resolution. It has 327 hand-written text-image-audio queries in 86 scene groups across 18 ambiguity categories, and removing any one modality makes the answer non-unique. Across 14 omni-modal systems, the best model trails humans by over 20 points in group accuracy. Without answer options, model group accuracy falls to single digits, and leave-one-out tests show audio cues are often underused.
evals, multimodal, research· Added 23 Oct 2026, published 22 Oct 2026
x.comthread🛡️ SecurityTrail of Bits post on two unpatched critical PHP bugs that were exploited within 17 hours of details leaking. Cloudflare began blocking attacks and Patchstack reported active exploitation. The bug was reported 11 August and fixed 71 days later, with a patch touching 402 files and over 11,000 lines. Trail of Bits announces two open-source plugins, patching-bugs and regression-hunt, from its Patch the Planet work.
AppSec, AI Security· Added 23 Oct 2026, published 21 Oct 2026
x.comarticle🧠 AITrail of Bits announces two open-source plugins, patching-bugs and regression-hunt, for AI coding agents that speed up bug patching. They come from four months of Patch the Planet work with open-source projects, which found that repo-level documentation, skills and plugins give agents the maintainer context on what counts as a bug, how a fix should look and what must not break. The post cites the October 2026 PHP incident: a patch touching 402 files, and exploitation within 17 hours of leaked details.
coding-agents, prompting· Added 23 Oct 2026, published 21 Oct 2026
blog.trailofbits.comarticle🧠 AITrail of Bits releases patching-bugs and regression-hunt, two agent plugins built with OpenAI through Patch the Planet. patching-bugs reproduces bugs, writes fixes and runs a three-arm proof; regression-hunt traces diffs to real interfaces, confirms with live PoCs and a fresh-context replay. A review of 130+ merged PRs found 21 regressions, and regression-hunt gave 25% fewer false positives than a baseline Codex prompt.
coding-agents, agents, evals· Published and added 21 Oct 2026
blog.trailofbits.comarticle🛡️ SecurityTrail of Bits releases patching-bugs and regression-hunt, two Codex plugins from the Patch the Planet initiative with OpenAI. A review of 130+ merged patches confirmed 21 regressions (none broke a security protection). In preliminary tests regression-hunt gave 25% fewer false positives than a baseline Codex agent. Design uses live PoCs, strict schemas, a three-arm proof, and project THREAT_MODEL.md and PATCHING_GUIDELINES.md files.
AI Security, AppSec· Published and added 21 Oct 2026
x.comarticle🛡️ SecurityCVE-2026-110936 is an uncontrolled DLL search path flaw (CWE-427) in Siemens Software Center below V4.0.9.8, listed in CISA KEV on 20 October 2026. The post covers CVSS v4.0 8.4 and v3.1 6.7, Siemens' Romulus ClickFix campaign indicators (IPs, domain, hashes, file leads) and a six-step SOC trail using Sysmon, EDR and network records. It separates delivery from DLL load and from compromise.
Blue Team & Detection, Threat Intel· Added 24 Oct 2026, published 21 Oct 2026
alterauth.comtool🧠 AIAlter is a privileged access management layer for AI agents (YC S25). It stores credentials in a vault, injects them per request through a proxy, gives each agent its own identity and scoped grants, and logs every call to a tamper-evident audit trail. One SDK call covers storage, refresh, policy and audit for 140 API providers, with LangChain, MCP and AWS integrations. Policy is default-deny down to the parameter, with human approval for risky calls.
ai-infra, agents· Added 22 Oct 2026, published 21 Oct 2026
x.comthread🛡️ SecurityTrail 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.
Blockchain· Added 22 Oct 2026, published 20 Oct 2026
x.comadvisory🛡️ SecurityCVE-2026-77932 affects DB-GPT through 0.8.2 (dbgpt-app). The /api/v1/chat/react-agent route accepts unauthenticated requests because get_user_from_headers returns an admin-role user, and the ReAct agent can reach unsandboxed Python and shell tools. Lasso reported root in the webserver container in a lab. The post gives CVSS 9.8 and 9.3, no verified fix (0.8.2 is still the latest release), no KEV match, no EPSS data, and a SOC hunting trail.
AI Security, AppSec, Blue Team & Detection· Added 23 Oct 2026, published 20 Oct 2026
securitylabs.datadoghq.comtool🛡️ Securitypathrunner is a modular offensive framework from Datadog Security Labs that automates exploitation of the AWS IAM privilege escalation paths catalogued in pathfinding.cloud and deployed in Pathfinding Labs. It has over 80 exploit modules (as of October 2026), identity import and switching, multiple payloads per module, and attacker infrastructure deployment with a credential listener. A workspace report lists created resources and expected CloudTrail events for defenders.
Cloud & Infra, Offensive Security, Blue Team & Detection· Added 23 Oct 2026, published 20 Oct 2026
x.comthread🛡️ SecurityDFIR_Radar weekly detection engineering digest covering 23 new and 19 updated rules across 6 GitHub repositories. It lists KQL rules for Azure Key Vault secret retrieval, MCP client User-Agent bulk operations and local Azure token theft, AWS CloudTrail analytics, Sigma rules for firewall and ETW tampering (decimal 17 vs hex 0x11), updated LOLDrivers hashes, and Sublime email rules for OAuth consent phishing and fake IRS W-9 PDFs.
Blue Team & Detection, Cloud & Infra, AI Security· Added 20 Oct 2026, published 19 Oct 2026
x.comadvisory🛡️ SecurityCVE-2026-106931 is an OS command injection (CWE-78) in the statistics processing of EG Secure Solutions SiteGuard Server Edition, a host-based WAF. It can give root or SYSTEM command execution. The vendor and JVN confirm exploitation on 14 October 2026. The post lists fixed builds (Ver8.00 Update 5, Ver7.00 Update 6 on FreeBSD), the IIS exclusion, the workaround of disabling statistics, and a SOC trail starting with diagnostic preservation before patching.
Blue Team & Detection, AppSec· Added 21 Oct 2026, published 19 Oct 2026
github.comrepo🧠 AIKYB (Know Your Business) is a Rust/axum server that stores agent-written knowledge, incident reports and tasks as Markdown files in Git, one commit per change. Search is hybrid: Tantivy BM25 fused by reciprocal rank with local multilingual-e5-small ONNX vectors, and it can query historical versions or "as of" a date. It ships a kyb CLI, a native MCP endpoint (Claude Code, Codex, Antigravity, Gemini CLI), an installable agent skill, token auth with per-write attribution trailers, secret scanning on writes, and a web console. It targets persistent shared operational memory for multi-agent, multi-machine fleets.
agents, mcp, coding-agents· Added 17 Oct 2026
x.comadvisory🛡️ SecurityCyberForge CVE of the Day #080 covers CVE-2026-99999 ("PwnEscape"), a Linux kernel IPv4 checksum-completion flaw. Crafted checksum metadata can trigger a deferred write into an already-validated IPv4 header, which allows host memory disclosure from a guest or microVM over virtio-net and TAP. It gives upstream fixed branches (5.10.272 to 7.2.11), Amazon Linux, Red Hat and Ubuntu status, CVSS 9.3, EPSS 0.211% and a SOC investigation trail. No IOCs were verified.
Cloud & Infra, Blue Team & Detection· Added 19 Oct 2026, published 17 Oct 2026
x.comthread🛡️ SecurityTrail of Bits post announcing research that tested 25 package managers against a simulated malicious repository serving swapped code for the same package and version. Without a lockfile only Go and vcpkg verified every download; with a lockfile fewer than half did. This undercuts the assumption that lockfiles make dependencies immutable and shows hash/integrity verification gaps across ecosystems.
Supply Chain· Added 18 Oct 2026, published 16 Oct 2026
x.comthread🛡️ SecurityCyberForge "CVE of the Day #079" on CVE-2026-102764, a CWE-290 origin-validation flaw in BeyondTrust Remote Support (advisory BT26-05, CVSS v4 9.3) that allows appliance file operations. It summarises Mandiant/GTIG's exploited intrusion chain (SAWFLY web shell, BARECLAW backdoor, root escalation, persistence) and gives an investigation trail: patch-state verification (BT26-05-RS), evidence preservation, hunting leads with SHA-256 hashes and YARA rules, permission/session correlation, and remediation including clean redeployment and credential rotation. It separates vendor facts from the intrusion's observed behaviour and notes that KEV and EPSS have no entries.
Blue Team & Detection, Threat Intel· Added 18 Oct 2026, published 16 Oct 2026
github.blogarticle🛡️ SecurityGitHub engineers describe Evergreen, a Go-based GitHub App that runs quarterly outside-collaborator access reviews and archives stale repositories in a shared org. It uses scheduled Actions with repository_dispatch, JSON state files in git as the audit trail, checklist issues (IssueOps) closed to trigger revocation, and custom properties for ownership and exemptions. Design principles include fail-secure/fail-closed revocation, never deleting data, dry-run rollout, Splunk logging, and proactive pacing against GitHub's secondary rate limits.
DevSecOps, Identity & Access· Published and added 15 Oct 2026
x.comthread🛡️ SecurityTrail of Bits tweet announcing a test of 25 package managers against a simulated malicious repository. Without a lockfile only Go and vcpkg verified every download; with a lockfile fewer than half did. Replies raise that checksum verification is separate from execution isolation, and ask whether poisoned CI package caches were in the threat model.
Supply Chain· Added 17 Oct 2026, published 15 Oct 2026
latent.spacearticle🛡️ SecurityRecap of AI Engineer New York talks on balancing agent autonomy and safety in financial services. Bloomberg's Andrey Rybka argued for deterministic controls outside the model: scoped credentials, tool-call checks, outbound traffic limits, audit trails, budgets and kill switches. He cited a case where an agent obeyed instructions embedded in a marketplace review. Anthropic described a trust layer of identity and permissions, sandbox and approval, and regression checks, with agents proposing rather than making order amendments. Two Sigma described composable controls, including auto-mode classifiers that escalate only important requests, to avoid approval fatigue. OpenAI framed human-in-the-loop as a product design choice.
AI Security· Published and added 14 Oct 2026
blog.trailofbits.comwriteup🛡️ SecurityTrail of Bits tested 25 popular package managers with a Docker and mitmproxy harness that simulates a compromised package repository injecting malware into legitimate packages. Fewer than half enforce lockfile hashes by default, only two verified every download, and only Go verifies signatures by default. The post covers lockfile hash pitfalls (Pub, pnpm, Conan, CocoaPods, Composer), repository, developer and Sigstore signing, and Go's checksum database. It also proposes lockfiles that commit to a signing identity plus a transparency log.
Supply Chain· Published and added 14 Oct 2026
embracethered.comwriteup🛡️ SecurityCVE-2026-103572 is a remote code execution bug in Microsoft.VisualStudio.Copilot.Core.dll, shared by Visual Studio and SQL Server Management Studio. A trailing dot (for example `.mcp.json.`) bypasses the protected-filename check in EditsFunctionHandler.GetEditConfirmationTypeAsync, so a hijacked Copilot can rewrite its own MCP and settings files and run commands. The post covers path traversal and InstallationID leaks, indirect prompt injection demos, the October 2026 patch (GetFullPath), and a $0 bounty.
AI Security, AppSec, Offensive Security· Added 29 Oct 2026, published 13 Oct 2026
arxiv.orgpaper🧠 AIIntroduces Sauron, an evidence-first agentic verifier that checks rubric criteria for long-horizon agent tasks by investigating both environment state and execution trace, and trusts a claim (including an agent's self-reported success) only after validating it against its source. Bounded queries and joint investigation of related criteria cut context use, and each verdict carries an auditable evidence receipt. On 2,586 human-labeled criteria from 100 personal-assistant episodes it has the best F1 of five verifiers across eight backbones (0.96 vs 0.92 for Gandalf with GPT-5.6 Sol, at 23% lower cost). It detects 97% of injected failures across 13 fault families, including reward-hacking simulations, with 2% false flags on clean controls, and leads on the TRAIL agent-error benchmark. It matters because unreliable rubric verdicts used as RL rewards can reinforce bad behavior.
evals, agents, research· Added 14 Oct 2026, published 13 Oct 2026
aws.amazon.comarticle🛡️ SecurityExplains how to turn a cryptographic algorithm policy (approve ML-DSA-65/87, exclude ML-DSA-44) into enforceable AWS controls using SCPs, IAM policies, permission boundaries and KMS key policies. The key lever is the kms:KeySpec condition key (kms:SigningAlgorithm can't distinguish variants, since all use ML_DSA_SHAKE_256), applied to CreateKey, ReplicateKey, Sign, Verify and GetPublicKey. It also covers surfaces without algorithm condition keys (AWS Private CA, ACM import, IAM server certificate upload, ELB TLS) via provisioning pipelines plus CloudTrail and AWS Config drift detection, and the default-allow risk of deny-lists.
Cloud & Infra, GRC· Published and added 12 Oct 2026
micahflee.comarticle🧠 AIMicah Lee argues for dropping OpenAI and Anthropic in favor of open-weight models, claiming they trail proprietary ones by about four months. They report GLM-5.3 as roughly on par with Claude Opus 5.5 for software engineering, and GLM-5.3-Flash as similar to Opus 4.8 and feasible to run locally. The post compares private inference through secure enclaves (Tinfoil, Maple AI) with cheaper non-private routers (OpenCode Go). It gives pricing, data-trust caveats about Z.ai, and the author's own workflow: OpenCode harness, GLM-5.3 on low reasoning, Tinfoil for private work.
models, inference, coding-agents· Published and added 12 Oct 2026
endorlabs.comarticle🧠 AIExplains how lifecycle hooks in Claude Code (.claude/settings.json) and Cursor (.cursor/hooks.json) can act as a deterministic, out-of-band control layer for AI coding agents. It covers the hook events, the JSON passed on stdin, and the response contracts. Claude Code uses exit codes plus permissionDecision, updatedInput and additionalContext. Cursor uses stdout JSON with permission and userMessage. It then shows how Endor Labs' `endorctl ai-audit` handler normalises events across both harnesses, streams them to a central audit trail and enforces server-side policies (29 defaults) such as shell allow/deny lists, sensitive-file patterns and MCP allowlists.
coding-agents, ai-policy· Published and added 12 Oct 2026
x.comadvisory🛡️ SecurityCyberForge's CVE of the Day #075 covers CVE-2026-108860, a critical (CVSS 9.1 v3.1 / 9.3 v4.0) authentication bypass in SciSharp BotSharp through 5.2.0. WebStarter ships a published default HMAC JWT signing secret (CWE-321), so anyone holding it can forge bearer tokens that impersonate known users, including administrators. The post sets out how to confirm the effective key and serving build, a six-step SOC investigation trail, and remediation by rotating the secret. No verified fixed release, KEV entry or EPSS data existed at cutoff, and the linked research is static-only.
AppSec, Identity & Access, Blue Team & Detection· Added 13 Oct 2026, published 12 Oct 2026
x.comtool🛡️ SecurityTrail of Bits announces SequenceHash, a library for hashing several inputs without concatenation ambiguity. Plain concatenation lets different input sequences produce the same SHA-256 digest. SequenceHash appends each input's length as a fixed 128-bit integer so every sequence has a single encoding. It ships for Rust, Go and Python, and removes a recurring bug class in crypto and appsec code.
AppSec· Added 11 Oct 2026, published 9 Oct 2026
decrypt.coarticle🧠 AIDecrypt reports on Google Cloud's newly announced Gemini Agent, a long-running AI agent that operates across Gmail, Docs, Sheets, Slides, Drive, Chat, Calendar, Microsoft 365 and Slack, and can run for hours or days on a single goal. The feature allows managers to spin up coworker agents with their own Workspace email address, calendar, Drive storage and directory listing, each given a cryptographically attested identity and a separate audit trail; agents connect to Salesforce, ServiceNow, Jira, Snowflake, BigQuery and any Model Context Protocol server, and are orchestrated across both Gemini and Anthropic's Claude models depending on cost and quality fit. Google pairs the launch with safety infrastructure, an Agent Sandbox, an AI network firewall called Agent Gateway, and real-time token-spend caps, citing a prior Claude Cowork VM-escape finding by Accomplish AI as the kind of containment failure it is designed to prevent.
agents, ai-infra· Published and added 8 Oct 2026
decrypt.coarticle🧠 AIAnthropic released Claude Haiku 5.5, pricing it at $0.10/$0.50 per million input/output tokens (for prompts up to 100k tokens), a 90% cut from Haiku 4.5, matching OpenAI's GPT-6 Luna pricing. The model scored 72.4% on OSWorld 2.1 and 39.2% on Terminal-Bench, beating GPT-6 Luna on both but trailing Claude Sonnet 5.5's 70.6% on Terminal-Bench, and it introduces an adjustable "effort" setting to trade cost for answer quality. Anthropic also halved Claude Sonnet 5.5's cache-read price and is rolling out monthly API credits for Max and Team subscribers.
models, evals· Published and added 7 Oct 2026
aws.amazon.comwriteup🧠 AIAWS Security Blog walkthrough showing how to build identity-aware AI data agents on Amazon Bedrock AgentCore that propagate a real user's OIDC id_token through the agent's tool-calling chain (Runtime → MCP Gateway → Lambda) via HTTP headers, keeping the token out of the foundation model's context entirely. A Lambda function performs a server-side token exchange using IAM Identity Center Trusted Identity Propagation, assumes a scoped TIP role, and runs the user's Athena query so AWS Lake Formation evaluates the real user's grants, producing per-user row/column access and CloudTrail audit records naming the actual human (onBehalfOf) without rewriting existing governance policies.
agents, mcp, ai-infra· Added 8 Oct 2026, published 6 Oct 2026
aws.amazon.comarticle🛡️ SecurityAWS Security Blog walkthrough of an identity-aware AI data agent pattern that propagates a real user's OIDC id_token through Amazon Bedrock AgentCore Runtime and Gateway into a Lambda target, which exchanges it server-side (via IAM Identity Center trusted identity propagation) for Lake Formation-scoped Athena credentials. The design deliberately keeps the id_token off the tool schema and inside HTTP headers only, so the foundation model never sees or handles the user's credential, and CloudTrail records the real human via an onBehalfOf AssumeRole entry, letting existing Lake Formation grants govern AI agent queries without rewriting authorization logic in application code. It details the three AgentCore configuration gates (runtime header allowlist, gateway metadataConfiguration.allowedRequestHeaders, Lambda's bedrockAgentCorePropagatedHeaders) that enforce this separation.
AI Security, Identity & Access, Cloud & Infra· Added 7 Oct 2026, published 6 Oct 2026
x.comthread🛡️ SecurityTrail of Bits' Circomspect, a static analyzer for Circom zero-knowledge circuits, had a false positive in its unconstrained-less-than check: compile-time constants fed into the LessThan template were wrongly flagged as unconstrained inputs. The merged PR fixes this so expressions resolvable at compile time are no longer reported, reducing noise for auditors hunting real under-constraint bugs in circuits.
AppSec· Added 8 Oct 2026, published 6 Oct 2026
mailaccess.protool🛡️ SecurityMailAccess is an open-source email OSINT framework (pip install mailaccess, MIT licensed, 1.6k★ on GitHub as KatrielMoses/MailAccess) that runs 75 modules across 5,300+ platforms to surface every account, breach, and exposed data point tied to a given email address without requiring API keys. Each finding is returned with its source, a confidence rating, and a one-sentence reasoning trail (e.g. username matches, avatar hash correlation) rather than as an unverified black-box hit. The project also offers a separate paid email-scraping service that extracts business contacts per domain.
OSINT· Published and added 6 Oct 2026
x.comwriteup🛡️ SecurityCyberForge CVE-of-the-Day #070 on CVE-2026-61500 in Rejetto HFS 3.0.0–3.2.0. The default Koa session-signing key came from Math.random() (CWE-338) and the unauthenticated login handshake leaked outputs of the same generator. That allows recovery of the key, a forged administrator session, and code execution through the legitimate server_code feature. It covers scoring (CVSS 9.8/9.3, EPSS, KEV status), the 3.2.1 fix, a six-step SOC hunting and remediation trail (loginSrp1/2, get_config/set_config pivots, COOKIE_SIGN_KEYS review), and offline jq/Python helpers for container and config metadata.
Blue Team & Detection, AppSec, Identity & Access· Added 9 Oct 2026, published 6 Oct 2026
x.comthread🧠 AIA 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.
evals, models· Published and added 5 Oct 2026
Showing 60 of 125
Show more