mcp-audit

Audit connected MCP servers for token overhead, redundancy, and security. Use when sessions feel slow or before adding new MCPs.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/mcp-audit/SKILL.md

Quality

Content

68%Weight 40%Scale 1-5

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

A well-structured, actionable audit procedure with concrete commands and thresholds. The main weakness is workflow validation: the destructive disable step lacks an explicit in-sequence verification checkpoint, capping workflow clarity at 3.

Suggestions

Add an explicit validation checkpoint in the step sequence before recommending disables, e.g. 'Step 5: Verify each disable candidate against `disabledMcpjsonServers` and confirm with the user before presenting.'

Provide a runnable command or one-liner to count tools per server programmatically rather than relying on manual estimation against the token bands.

Tighten the Key Insight section by trimming explanatory phrasing like 'regardless of whether you use those tools' to improve token efficiency.

Dimension	Reasoning	Score
Conciseness	Largely lean with sectioned structure and genuinely useful quantitative specifics (per-tool-count token estimates) Claude would not precisely know; minor trims possible such as 'regardless of whether you use those tools', so it sits above 3 but below the every-token-earns-its-place 5 anchor.	4 / 5
Actionability	Provides copy-paste bash commands in Step 1, concrete token bands in Step 2, explicit keep/disable rules, and a fill-in output template; counting tools is left as manual estimation rather than a runnable script, a minor gap below the fully copy-paste 5 anchor.	4 / 5
Workflow Clarity	Four steps are clearly sequenced (list -> count -> check usage -> recommend), but the destructive 'disable servers' action only has an implicit confirmation gate buried in the Rules section rather than an explicit validation checkpoint in the step flow; per the destructive-operation cap, this holds at 3 and below the 4 anchor which requires explicit checkpoints.	3 / 5
Progressive Disclosure	No bundle files exist and the content is self-contained with clean section headers (Trigger, Audit Steps, Output, Thresholds, Rules) and easy navigation; the file is over 50 lines so it does not fully qualify for the simple-skill 5 exception, placing it at good-structure 4 rather than 5.	4 / 5
	Total	15 / 20 Passed

Description

87%Weight 40%Scale 1-5

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

A strong, well-scoped description that clearly states the capability and concrete trigger conditions in third person. Minor room to broaden trigger synonyms, but it cleanly answers both what and when.

Dimension	Reasoning	Score
Specificity	Names the domain ('connected MCP servers') and three concrete audit dimensions ('token overhead, redundancy, and security'); one verb with multiple specific targets rather than many distinct actions, so it sits above the 3 anchor but below the comprehensive multi-action 5 anchor.	4 / 5
Completeness	Explicitly answers both what ('Audit connected MCP servers for token overhead, redundancy, and security') and when ('Use when sessions feel slow or before adding new MCPs') with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Natural triggers like 'sessions feel slow' and 'before adding new MCPs' cover the main use cases well, though synonyms like 'expensive' or 'context window full' are missing, keeping it just below comprehensive 5.	4 / 5
Distinctiveness Conflict Risk	MCP-server auditing is a clear niche with distinct, MCP-specific triggers; minimal overlap risk with other skills.	5 / 5
	Total	18 / 20 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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: rohitg00/pro-workflow
Path: skills/mcp-audit/SKILL.md
Commit: 7f7209d

Reviewed: 3 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.