switchboard
N

NVIDIA SkillSpector

Featured

by NVIDIA

Scan agent skills, plugins, and MCP servers before installing them — detects prompt injection, data exfiltration, malicious patterns, and supply-chain risk in Claude Code and Codex skills.

4
Skills
None
Auth
No
Streaming
No
Push

Skills

Skill Scanning

Analyzes an agent skill or plugin package for malicious instructions and risky capabilities before installation.

Injection Detection

Flags prompt-injection and instruction-override patterns embedded in skill files, docs, and tool descriptions.

Exfiltration Checks

Identifies network calls and file access paths that could move credentials or source code off the machine.

Supply Chain Audit

Inspects declared dependencies and install hooks for tampering and untrusted sources in the skill supply chain.

SecurityCode & DevToolsagent-securityskill-scanningprompt-injectionsupply-chain-securitystatic-analysisclaude-codemcp-security
Visit Agent

Related Agents

O

OpenSandbox

Runs AI-agent workloads in isolated Docker or Kubernetes sandboxes, exposing sandbox lifecycle, command, filesystem, an…

Infrastructure & OpsCode & DevTools
APIMCPCLI
5 skillsAPI Key
p

promptfoo

Tests and red-teams LLM apps, agents and RAG pipelines from declarative config, scanning for prompt injection, jailbrea…

SecurityCode & DevTools
CLIAPI
3 skillsNone
I

IDA Pro MCP

Bridges IDA Pro to LLM clients over MCP, exposing decompilation, disassembly, cross-references, type and stack edits, m…

SecurityCode & DevTools
MCPCLI
5 skillsNone
S

Snyk

Developer security platform with AI-powered vulnerability detection, fix suggestions, and automated security testing ac…

SecurityCode & DevTools
APIMCPCLI
0 skillsAPI Key
nvidia-skillspector
Scan agent skills, plugins, and MCP servers before installing them — detects prompt injection, data exfiltration, malicious patterns, and supply-chain risk in Claude Code and Codex skills.
fields
nameNVIDIA SkillSpector
providerNVIDIA
urlhttps://docs.nvidia.com/skills/scanning-agent-skills
categoriessecurity · code-devtools
accesscli
authnone
streamingfalse
pushfalse
verifiedtrue
tagsagent-security, skill-scanning, prompt-injection, supply-chain-security, static-analysis, claude-code, mcp-security
skills
skill-scanningSkill ScanningAnalyzes an agent skill or plugin package for malicious…
injection-detectionInjection DetectionFlags prompt-injection and instruction-override pattern…
exfiltration-checksExfiltration ChecksIdentifies network calls and file access paths that cou…
supply-chain-auditSupply Chain AuditInspects declared dependencies and install hooks for ta…
→ https://docs.nvidia.com/skills/scanning-agent-skills