For securing agent skill files, the most directly relevant products are dedicated skill and MCP scanners: SkillsSafe scans SKILL.md, MCP configs and system prompts for credential theft, exfiltration and prompt injection 9, and SkillTotal statically scans MCP servers and agent skills for prompt injection and tool poisoning 10. Broader application security platforms that also cover skill-related checks include Semgrep, which scans source code for vulnerabilities with SAST, SCA and secrets dete
- 1
Sentrysentry.ioverified and signed
Its skill-scanner action scans agent skills for security issues, including checks for injection, per its action record 11.
official:MCP server · @sentry/mcp-serverSDK · @sentry/nodeCLI · sentry-clirepository · sentry-pythonskill · sentry-cli+44 community tools built for it
for agents: skill-scanner · skill-writer · sentry-sdk-skill-creator
- 2
Semgrepsemgrep.devverified and signed
Its security platform scans source code with AI-assisted SAST, SCA and secrets detection to surface real vulnerabilities 6, and its code-security and llm-security skills give secure-coding and LLM security guidance 16.
+2 community tools built for it
for agents: code-security · llm-security · semgrep
- 3
SkillsSafe Security Scannerskillssafe.comwebsite read, tools not yet checked
A free scanner that checks SKILL.md files, MCP configs and system prompts for credential theft, data exfiltration, shell injection and prompt injection before install 9.
official:MCP server · SkillsSafe Security Scanner+1 community tools built for it
- 4
SkillTotalskilltotal.aiwebsite read, tools not yet checked
A free static scanner for MCP servers and agent skills that finds prompt injection, tool poisoning and exfiltration with file:line evidence 10.
official:MCP server · SkillTotal+1 community tools built for it
- 5
GitHubgithub.comverified and signed
Its mcp-security-audit skill audits MCP server configurations such as .mcp.json files for security issues 12.
official:CLI · github-mcp-serverskill · add-community-extensionskill · git-commitskill · acquire-codebase-knowledge+2,468 community tools built for it
for agents: security-review · mcp-security-audit · make-skill-template
Sources
- 1Sentry Docs | Application Performance Monitoring & Error Tracking Softwaredocs.sentry.io · read 2026-10-29
- 2GitHub · Change is constant. GitHub keeps you ahead. · GitHub · GitHubgithub.com · read 2026-10-29
- 6Semgrep App Security Platform | AI-assisted SAST, SCA and Secrets Detectionsemgrep.dev · read 2026-10-29
- 9SkillsSafe — Free AI Skill Security Scanner | Check Before You Installskillssafe.com · read 2026-10-29
- 10MCP Server & Agent Skill Security Scanner · SkillTotalskilltotal.ai · read 2026-10-29
- 11skill-scanner: declares "skill-scanner" (the record we read and signed)unfragile.ai · read 2026-10-24
- 12security-review: declares "security-review" (the record we read and signed)unfragile.ai · read 2026-10-24
- 16code-security: declares "code-security" (the record we read and signed)unfragile.ai · read 2026-10-24
For agents: the 10 actions behind this answer
skill-scanner · skill-scanner verified and signed
Scan agent skills for security issues. Use when asked to "scan a skill", "audit a skill", "review skill security", "check skill for injection", "validate SKILL.md", or assess whether an agent skill is safe to install. Checks for prompt injection, malicious scripts, excessive permissions, secret expo
security-review · security-review verified and signed
AI-powered codebase security scanner that reasons about code like a security researcher — tracing data flows, understanding component interactions, and catching vulnerabilities that pattern-matching tools miss. Use this skill when asked to scan code for security vulnerabilities, find bugs, check for
claude-security/skill/claude-security · claude-security verified and signed
Skill claude-security (in the claude-security plugin)
create-skill · create-skill verified and signed
Scaffolds new agent skills for the dotnet/skills repository. Use when creating a new skill, generating SKILL.md files, writing a skill description that the runtime will actually route to, or setting up skill directory structures. Handles frontmatter generation, section templates, and validation guid
docker-sandboxes-kits · docker-sandboxes-kits verified and signed
Use this skill when authoring, validating, packaging, signing, or composing a Docker Sandboxes kit `spec.yaml` (`sbx kit add/inspect/pack/pull/push/sign/validate/verify`), even if the user just says they want to "add a tool to a sandbox agent", "build a reusable sandbox extension", "publish a kit to
code-security · code-security verified and signed
Security guidelines for writing secure code. Use when writing code, reviewing code for vulnerabilities, or asking about secure coding practices like 'check for SQL injection' or 'review security'. IMPORTANT: Always consult this skill when writing or reviewing any code that handles user input, authen
Read · appsec-agent found in code, not yet checked
Read the UTF-8 contents of a file within the project working directory.
get_files_id_metadata_global_boxSkillsCards · Box Platform API verified and signed
List Box Skill cards on file
· SkillsSafe website read, tools not yet checked
Free security scanner for AI agent skills. Scan any SKILL.md, MCP config, or system_prompt for credential theft, data exfiltration, and prompt injection. Supports OpenClaw, Claude Code, Cursor, Codex. No signup required.
· SkillTotal website read, tools not yet checked
Free static scanner for MCP servers, agent skills and npm/PyPI packages. Finds prompt injection, tool poisoning and exfiltration with file:line evidence.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.