Skills

This skill facilitates silent installation of arbitrary external skills

This skill enables arbitrary command execution via hook scripts and

The skill misrepresents itself as a

No security issues detected in anthropics/skills/frontend-design.

This skill exposes sensitive credentials, allows privilege escalation, arbitrary

The skill executes arbitrary remote content from mutable, unauthenticated

The `azure-deploy` skill relies on custom, unaudited 'MCP Tools', posing a supply chain risk due to unknown

The skill exposes Azure CLI commands

The skill's human approval step for

The skill is vulnerable to command and K

No security issues detected in microsoft/azure-skills/azure-compute.

The skill introduces supply chain risks through external dependencies and local file loading, potentially influencing agent reasoning if compromised.

No security issues detected in microsoft/azure-skills/azure-messaging.

No security issues detected in microsoft/azure-skills/azure-hosted-copilot-sdk.

No security issues detected in microsoft/azure-skills/appinsights-instrumentation.

No security issues detected in microsoft/azure-skills/entra-app-registration.

The skill introduces supply chain risks and indirect prompt injection

The skill misrepresents its capabilities, claiming full

This skill is vulnerable to prompt injection, allowing

The skill enables reconnaissance of sensitive Azure Key Vault artifacts by listing and retrieving metadata for keys, secrets,

The skill risks command injection via `az graph query`

This skill allows command injection, data exfiltration,

The skill allows powerful Azure resource management and sensitive data querying

The skill enables direct command execution via Azure CLI, posing a risk for arbitrary command execution if the agent's environment is not properly sandboxed.

The skill risks command injection and arbitrary file system access due to unsanitized FFmpeg inputs.

This skill performs Azure resource reconnaissance and administrative actions, posing

The skill can execute arbitrary code, deploy malicious resources,

The skill enables arbitrary command execution, data exfiltration, and social engineering, while also introducing supply chain vulnerabilities through dynamic skill loading.

This skill can provision, modify, destroy, and

The skill performs cloud reconnaissance and is vulnerable to supply chain

This skill can poison agent knowledge bases through malicious injection into

The skill allows command injection, poisons the

This self-improving skill autonomously modifies critical

No security issues detected in vercel-labs/agent-skills/vercel-composition-patterns.

The skill allows arbitrary Azure API calls and

The skill misrepresents its Azure service support,

This skill executes powerful cloud deployment commands that can cause significant infrastructure changes or destruction if misused.

The skill enables arbitrary file operations and Azure

No security issues detected in microsoft/github-copilot-for-azure/azure-diagnostics.

No security issues detected in microsoft/github-copilot-for-azure/entra-app-registration.

This skill executes arbitrary commands from external files, manipulates

The skill uses powerful Azure CLI commands and handles

The skill is vulnerable to command

No security issues detected in microsoft/github-copilot-for-azure/azure-rbac.

The skill allows arbitrary KQL query execution and reconnaissance

No security issues detected in microsoft/github-copilot-for-azure/azure-prepare.

No security issues detected in microsoft/github-copilot-for-azure/azure-compliance.

The skill deceptively provides actionable Azure CLI and Bicep commands for resource creation and code modification, contradicting its stated purpose.