mcp-server-review

Security review of MCP (Model Context Protocol) server implementations and configurations. Use when auditing MCP server source code, evaluating third-party MCP servers before installation, or reviewing Claude Code MCP integrations for overpermissioning, injection risks, and data exposure.

Quality

78%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/mcp-server-review/SKILL.md

Quality

Discovery

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This is an excellent skill description that clearly defines a specific niche (MCP server security review), lists concrete actions and risk categories, and includes an explicit 'Use when' clause with multiple natural trigger scenarios. It uses proper third-person voice throughout and would be easily distinguishable from other skills in a large skill library.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: 'auditing MCP server source code', 'evaluating third-party MCP servers before installation', 'reviewing Claude Code MCP integrations for overpermissioning, injection risks, and data exposure'. These are clear, actionable capabilities.	3 / 3
Completeness	Clearly answers both 'what' (security review of MCP server implementations and configurations) and 'when' with an explicit 'Use when' clause covering three distinct trigger scenarios: auditing source code, evaluating before installation, and reviewing integrations for specific risk categories.	3 / 3
Trigger Term Quality	Includes strong natural trigger terms: 'MCP', 'Model Context Protocol', 'security review', 'auditing', 'MCP server', 'third-party MCP servers', 'overpermissioning', 'injection risks', 'data exposure', 'Claude Code MCP integrations'. Good coverage of terms a user concerned about MCP security would naturally use.	3 / 3
Distinctiveness Conflict Risk	Highly distinctive niche combining security review with the specific domain of MCP servers. The combination of 'MCP', 'security', 'overpermissioning', and 'injection risks' creates a clear, unique trigger profile unlikely to conflict with general security or general MCP skills.	3 / 3
	Total	12 / 12 Passed

Implementation

57%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill is well-structured and concise, covering MCP security review comprehensively with specific, non-obvious checks that genuinely augment Claude's capabilities. However, it suffers from referencing external files that don't exist in the bundle, lacks executable examples or concrete output templates, and is missing validation checkpoints that would be expected in a security audit workflow.

Suggestions

Provide the referenced bundle files (plays/tier4-ai-security/mcp-server-review.md and templates/finding.md) or inline their essential content—currently these are dead references that undermine the skill's usability.

Add a concrete example of a tool risk matrix row and a sample finding entry so Claude knows the exact expected output format.

Add validation checkpoints, e.g., 'After Step 2, confirm all tools are catalogued before proceeding to injection analysis' and 'Cross-reference findings against tool classifications before generating the report.'

Dimension	Reasoning	Score
Conciseness	Every line serves a purpose. No unnecessary explanations of what MCP is or how security reviews work in general. The checklist items are dense with specific, actionable checks (e.g., '169.254.169.254' for cloud metadata, '../' for path traversal) that add genuine value beyond Claude's baseline knowledge.	3 / 3
Actionability	The skill provides specific things to check (command injection via shell interpolation, SSRF, path traversal) and concrete classification categories, but lacks executable code examples—no sample commands for SCA, no code snippets showing what vulnerable patterns look like in source, no example tool risk matrix or finding template content. It references external files (plays/tier4-ai-security/mcp-server-review.md, templates/finding.md) but these aren't provided.	2 / 3
Workflow Clarity	The 7 steps provide a clear sequence for the review process, but there are no validation checkpoints or feedback loops. For a security audit (which can have destructive recommendations and involves careful verification), there's no step for verifying findings, no guidance on what to do when issues are found mid-review, and no explicit checkpoint before producing the final output.	2 / 3
Progressive Disclosure	The skill references 'plays/tier4-ai-security/mcp-server-review.md' and 'templates/finding.md' but no bundle files are provided, making these dead references. The skill delegates the 'full procedure' to an external file while also inlining what appears to be the full procedure, creating confusion about what the referenced file adds.	1 / 3
	Total	8 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: OWASP/secure-agent-playbook
Commit: 3f4fcb6

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.