Claude Code Guide
OCT 2026 Latest: Mistral Large 4 AI FinOps Lean & AI Changelog →

Security hub

Secure Claude Code and AI Agents

Start with a local repository check, review current evidence, then follow the security path that matches your task.

Threat DB v2.31.0Updated October 4, 2026

Check a repositoryChoose a path

Local check

Start with your repository

Install AgentSec and run its current alpha scanner locally. This website does not receive or inspect your files.

AgentSec 0.1.0a0
agentsec scan /path/to/repository --format json --redact

Current evidence, with its date

These values come from the bundled AgentSec feed and describe that database snapshot. They are not live measurements of the entire ecosystem.

Skills Scanned
3,984
Have Flaws
36.82%
Critical-Risk
534
Malicious Payloads
76
CVEs & Advisories
184
Exposed Servers
~1,000

Verify your repository with AgentSec

The scanner checks bounded repository inputs against bundled intelligence. A clean result is not proof of safety.

2. Run locally

Execute the scan inside the target repository. Review the exact files and evidence behind each finding.

Inspect AgentSec source →

3. Respect the boundary

Current detectors do not cover every host-level persistence mechanism, Git-history witness, or runtime behavior.

2 detector records
  • clawhavoc-skill v1: Detect repository-local agent-skill evidence associated with the documented ClawHavoc fake-prerequisite campaign.
  • shai-hulud-keyv v1: Detect repository-local indicators associated with the August 2026 Shai-Hulud Keyv/cacheable npm campaign.

Choose the task, not the scroll depth

Each route answers one intent completely. The hub keeps orientation and current evidence without repeating every catalogue.

Threat intelligence

Understand how current attacks work, which ecosystems they affect, and which evidence supports each record.

Open this path

Vulnerability database

Check whether a component is affected, inspect the primary source, and find the documented remediation.

Open this path

Isolation guide

Choose an isolation boundary, understand what it excludes, and verify the controls before running an agent.

Open this path

Defensive baseline

Reduce immediate risk, strengthen a workstation, and establish repeatable controls for a development team.

Open this path

Latest sourced intelligence

Three recent feed events, ordered by their canonical date. Follow the source before making a time-sensitive decision.

Cowork macOS file-open host execution

An omitted executable file type allowed content from a Cowork shared folder to run commands on macOS when opened from Claude Desktop. A separate guest-kernel flaw could remove the user-interaction requirement on older VM images.

Anthropic

MCP Python client external schema reference fetch

Tool output-schema validation resolved server-chosen external references through synchronous URL or local-file reads without a timeout. Retrieved content was used as a schema and was not returned to the server; a stalled fetch could block the event loop.

Model Context Protocol maintainers

MCP Python OAuth authorization-server credential confusion

An untrusted MCP server could select where an OAuth client sent credentials because issuer validation and credential binding were incomplete. stdio clients, clients attaching their own headers and SDK servers are excluded.

Model Context Protocol maintainers

Open all threat intelligence →

Three checks before you continue

Use this quick pass now. The full hardening route keeps the complete workstation and team checklist.

0 of 3 complete

Repository
Runtime
Team controls

Open the complete checklist →

Looking for a section from the previous page?

The content remains available at a focused destination. These anchors preserve old bookmarks and inbound links.

Attack techniques

Mechanisms, examples, campaigns, and mitigations.

Open threats →

CVE database

Searchable vulnerabilities, fixed-version status, and mitigations.

Open CVEs →

AgentSec intelligence

Dated event records and explicit detector coverage.

Open intelligence →

Active campaigns

Tracked campaign records with reported dates and sources.

Open campaigns →

Threat database

Filter events, campaigns, authors, and malicious skills.

Search threats →

The sandbox

Protection boundaries, silent failures, configuration, and runtime checks.

Configure sandbox →

Defense tools

Capabilities and limits of the maintained scanner collection.

Compare tools →

Built-in commands

Quick checks, full audits, and the threat database terminal.

Open commands →