HOL LogoGuard

Explore HOL

  • HOL home
  • AI agent registry
  • AI plugins
  • Open standards
  • HOL members
  • Sponsor HOL
  • Advertise with HOL

Guard product

  • Guard overviewLocal security and control for AI agents and the tools they use.
  • FeaturesRuntime protection, policy routing, review, and evidence.

Explore Guard

  • Product previewWalk through Guard surfaces in read-only demo mode.
  • ComparisonCompare Guard with native controls and AI security vendors.

AI tools

  • All AI toolsEvery supported AI tool and how Guard applies policy to it.
  • Codex
  • Claude Code
  • Cursor
  • Antigravity CLI
  • OpenCode
  • Hermes
  • OpenClaw
  • GitHub Copilot CLI
  • Antigravity
  • Kimi
  • Grok
  • Pi / Oh My Pi
  • Zcode

Extensions

  • All extensionsBrowse command and MCP coverage with owners and stated limits.
  • Command coverageShell command protection across clouds, databases, backups, and packages.
  • MCP server coverageSee how Guard maps risk state across MCP tools and servers.
  • Core safetyThe safety floor listings that ship with Guard.
  • Data and resilienceBackup and storage command protection.
  • Cloud and infrastructureAWS, Azure, GCP, Kubernetes, and more.

Security

  • AI security hubSecurity research, advisories, and agent safety coverage.
  • AI tool securitySecurity profiles for each supported coding agent.
  • Safe labsHands-on attack simulations with safe boundaries.
  • Redacted warningsReal blocked actions with sensitive details removed.
  • AdvisoriesCoordinated disclosure reports for AI tooling.
  • Active CVEsSearch active CVEs affecting AI tooling.

Learn

  • Security guidesPractical guides for securing AI agent workflows.
  • DocsInstall, configure, and operate Guard with confidence.
  • ResearchPublished security research, benchmarks, and methodology.

Community

  • ReleasesVersion history, shipped changes and upgrade notes.
  • ContributorsThe people and contributions behind HOL Guard.
  • AffiliatesShare Guard with your audience and earn from referrals.
  • SponsorFund agreed research, implementation and shared testing through the HOL Partner Program.
  • AdvertiseCompare README logo placements, previews and monthly prices, or promote a plugin or extension.
PricingEnterpriseOpen AppInstall Guard
  1. Guard
  2. Security
  3. CVEs
  4. CVE 2026 62847 metalama allows local code execution through dll
HOL Guard

Public security guidance for teams protecting AI harnesses, MCP servers, skills, prompts, and local tool execution.

Install Guard

AI Security

  • Prompt injection
  • MCP security
  • OWASP MCP mapping
  • Supply chain

Resources

  • Trust packet
  • Harness setup
  • Redacted warnings
  • Safe labs

Product

  • Install Guard
  • Pricing
  • Open dashboard
Guard
  • Guard Overview
  • Releases
  • Contributors
  • Install Guard
  • Pricing
Docs
  • Documentation Index
  • Developer Hub
  • API Reference
  • Root OpenAPI
  • Registry OpenAPI
  • Run in Postman
  • Standards
  • Submit ERC-8004 Contract
  • Feature Your Agent
Best Plugins
  • Browse Plugins
  • Plugin Launches
  • Best Claude Plugins
  • Best Codex Plugins
  • Best Grok Plugins
  • Best Kimi Plugins
  • Best DeepSeek Plugins
  • Best Antigravity Plugins
  • Best MCP Servers
  • Best Cursor Plugins
  • Best OpenCode Plugins
Best Agents
  • Best ERC-8004 Agents
  • Best Virtuals Agents
  • Best MCP Servers
  • Best A2A Agents
  • Best x402 Payable
  • All Categories
Community
  • Telegram
  • X
More
  • About HOL
  • Sponsor
  • Advertise
  • Contact
  • Blog
  • GitHub
  • Privacy
  • Terms of Service
Settings

Copyright © 2026 HOL DAO LLC. All rights reserved.

Back to active CVEs
Medium · CVSS 8.2CVE-2026-62847GHSA-M6WQ-39M7-9XPP

Metalama allows local code execution through DLL planting in shared temporary directoriesCVE-2026-62847

Answer in brief

CVE-2026-62847 records a Medium severity (CVSS 8.2) vulnerability in Metalama allows local code execution through DLL planting in shared temporary directories. The current sources do not mark it as known exploited. The current feed maps Metalama.Framework (nuget), Metalama.Framework (nuget), Metalama.Framework (nuget), Metalama.Tool (nuget) and additional mapped packages. Check affected ranges and fixed versions before updating.

Analysis pending evidence review

HOL Guard separates source facts from reviewed analysis. See the methodology.

Published Oct 14, 2026Updated Oct 14, 2026Source checked Oct 14, 2026First seen by HOL Oct 14, 2026Material review Oct 14, 2026
Upstream Advisory

Key facts

Risk
Medium · CVSS 8.2
Exploitation
Not marked as known exploited
Affected software
6 mapped packages or products
Fix availability
Available

Why this deserves its current priority

CVSS is 8.2. The current sources do not mark it as known exploited. Treat this as a source-backed prioritization signal, not a statement about your environment.

Analysis status

Analysis pending evidence review

Factual feed record only; HOL analysis is not approved for indexing. Read the methodology.

Affected scope and exposure questions

The current feed maps Metalama.Framework (nuget), Metalama.Framework (nuget), Metalama.Framework (nuget), Metalama.Tool (nuget) and additional mapped packages. Check affected ranges and fixed versions before updating.

Mapped affected packages and fixed versions
PackageAffected rangeFixed version
Metalama.Frameworknuget>=2026.1.0-preview,<2026.1.192026.1.19
Metalama.Frameworknuget<2025.1.182025.1.18
Metalama.Frameworknuget>=2026.0.0-preview,<2026.0.252026.0.25
Metalama.Toolnuget>=2026.1.0-preview,<2026.1.192026.1.19
Metalama.Toolnuget<2025.1.182025.1.18
Metalama.Toolnuget>=2026.0.0-preview,<2026.0.252026.0.25

Recommended response

  1. 1Check inventory. Check lockfiles and deployed manifests for Metalama.Framework, Metalama.Framework, Metalama.Framework.
  2. 2Review the reported fix. Update Metalama.Framework to 2026.1.19; Metalama.Framework to 2025.1.18; Metalama.Framework to 2026.0.25; Metalama.Tool to 2026.1.19; Metalama.Tool to 2025.1.18; Metalama.Tool to 2026.0.25 if you use the affected versions. Test the change in a non-production environment first.

Evidence timeline and material changes

  1. Published upstream

    Oct 14, 2026

    Evidence: source:ghsa:source_dates:source-dates:record
  2. Source modified

    Oct 14, 2026

    Evidence: source:ghsa:source_dates:source-dates:record
  3. First seen by HOL

    Oct 14, 2026

Sources and claim methodology

  • GitHub security advisorygithub.com
  • GitHub security advisorygithub.com
  • GitHub security advisorygithub.com
  • GitHub security advisorygithub.com
  • GitHub security advisorygithub.com
  • GitHub security advisorygithub.com
Upstream source description

## Summary Metalama loaded and executed compile-time assemblies, auxiliary executable tools, and embedded helper assemblies from predictable temporary directories. On Linux and macOS the default cache location was the world-writable system temporary directory (`Path.GetTempPath()`, i.e. `/tmp`), and previously-generated artifacts were trusted based on the presence of a sentinel/`.completed` marker file rather than on any integrity verification. A local, unprivileged user on a shared machine or CI/build agent can pre-create the deterministic cache directory structure and plant a malicious DLL together with the matching sentinel file. When a higher-privileged user (or build/service account) subsequently runs a build, Metalama finds the sentinel, skips regeneration, and loads and executes the attacker-supplied assembly inside the compiler / IDE / MSBuild process. Affected load-and-execute paths: - the cache of compile-time assemblies that Metalama generates from your aspects; - auxiliary executable tools that Metalama extracts to a local directory and then launches; - embedded .NET helper assemblies that Metalama extracts to a local directory and loads into the build / IDE process. ## Impact Arbitrary code execution in the security context of the building user — i.e. a local privilege escalation across user accounts on shared POSIX systems (multi-user developer machines, CI/build agents). The attacker obtains code execution as the victim without holding the victim's privileges, and can read source/secrets, tamper with build output (supply-chain injection), or disrupt builds. ## Affected configurations - **Linux and macOS, default configuration — affected.** The shipped default uses the world-writable system temp directory. - **Windows, default configuration — not affected.** The default per-user temporary directory is protected by OS ACLs, so cross-user planting is not possible. - The `METALAMA_TEMP` environment variable is a deployer-controlled override and is **out of scope**: pointing it at an insecure (e.g. world-writable or attacker-owned) location is a deployment misconfiguration, not a defect in the product. ## Severity CVSS v3.1: `AV:L/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H` = **8.2 (High)** for the affected POSIX default configuration (Scope:Changed reflects the cross-user privilege escalation). CVSS v4.0 cross-check: `CVSS:4.0/AV:L/AC:L/AT:P/PR:L/UI:P/VC:L/VI:H/VA:L/SC:L/SI:H/SA:H` = 6.0 (Medium), which more granularly attributes low impact to the cache itself and high impact to the subsequent build process. The honest range is therefore Medium-High depending on the scoring model and platform. ## Fix Caches and extraction directories were relocated from the world-writable system temp directory to per-user application-data locations (`$XDG_CACHE_HOME` / `~/.cache/Metalama` on POSIX), relying on OS-enforced per-user ownership as the trust boundary. Fixed in **2026.1.19**. ## Weaknesses - CWE-377: Insecure Temporary File - CWE-427: Uncontrolled Search Path Element ## Affected packages - Metalama.Framework < 2026.1.19, - Metalama.Tool < 2026.1.19 Internal tracking: metalama/Metalama#1650.

Quoted source text, attributed separately from HOL analysis.

Record context

Vulnerability class
RCE
EPSS
Not reported
CWE IDs
CWE-377, CWE-427
Source
GitHub Security Advisories
Source checked
Oct 14, 2026
References
6 linked sources
Open source record