CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

aws-mcp-server

How to use the AWS MCP Server effectively — the single authenticated endpoint that lets an agent run AWS CLI/API calls, execute sandboxed Python, search and read live AWS documentation, and retrieve skills. Use when configuring the AWS MCP Server, deciding between a scripted run and a direct call, paginating list operations, working across multiple regions or profiles, or writing IAM policies that scope what the agent may do (aws:ViaAWSMCPService and aws:CalledViaAWSMCP condition keys) with CloudWatch and CloudTrail visibility.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

78%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-organized, lean instructional skill with concrete commands, exact flags, and clear multi-step working patterns with validation checkpoints. It falls short of excellent on conciseness (time-sensitive version guidance outside a deprecated section), actionability (no worked example of an actual call or pagination script), and workflow clarity (no failure-recovery loop).

Suggestions

Add one short worked example of a common case — e.g., a minimal sandboxed Python snippet showing a list → filter → describe run with pagination handling — to close the actionability gap for the most frequent usage pattern.

Add a brief error-recovery note to the working patterns (e.g., what to do when an API call returns an error or a pagination token never terminates) to give the workflow a validation feedback loop.

Move or frame the version-specific pinning example (mcp-proxy-for-aws@1.6.3) as a non-authoritative illustration and drop the 'check PyPI periodically' padding to tighten conciseness.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence (no explanations of what AWS or MCP is), but contains time-sensitive version information ('mcp-proxy-for-aws@1.6.3', 'check PyPI periodically') outside any deprecated/old-patterns section, plus slight conceptual repetition of the description's auth/observability framing — minor instances that could be trimmed.

4 / 5

Actionability

Mostly executable guidance: the proxy command 'uvx mcp-proxy-for-aws@<version> https://aws-mcp.us-east-1.api.aws/mcp', the '--metadata AWS_REGION=<region>' flag, 'aws configure' or 'aws sso login', and exact condition keys. Minor gaps: no example tool invocation or sandboxed-Python snippet for the common cases (a first API call, a pagination loop), even though the instruction-only format itself is not penalized.

4 / 5

Workflow Clarity

Working patterns are clearly sequenced with most checkpoints present — 'List then describe... never infer state from names alone' and 'Always paginate... page through every page before reporting counts or conclusions' are explicit validation checkpoints. However, there is no error-recovery guidance (what to do when an API call or script fails), leaving a minor validation gap.

4 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent) and none are needed at this size: the ~60-line body is split into five well-organized, clearly headed sections (What it exposes, Authentication model, Working patterns, Governing, Version pinning) with nothing that clearly belongs in a separate file.

5 / 5

Total

17

/

20

Passed

Description

92%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 description: comprehensive and specific about capabilities, third-person, with an explicit multi-scenario 'Use when' clause and distinct niche triggers. The only weakness is minor — a handful of natural trigger phrasings (credential setup, audit requests) are not covered.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 'run AWS CLI/API calls, execute sandboxed Python, search and read live AWS documentation, and retrieve skills' — comprehensively covering the server's four capability areas with no padding or filler.

5 / 5

Completeness

Both 'what' ('the single authenticated endpoint that lets an agent run AWS CLI/API calls, execute sandboxed Python, search and read live AWS documentation, and retrieve skills') and 'when' (an explicit 'Use when' clause listing five concrete trigger scenarios) are clearly and explicitly answered.

5 / 5

Trigger Term Quality

Strong natural trigger phrases: 'paginating list operations', 'working across multiple regions or profiles', 'writing IAM policies', 'CloudWatch and CloudTrail visibility'. A few natural user phrasings are missing (e.g., credential/SSO setup, 'truncated results', auditing agent activity), which keeps it just below the comprehensive-with-synonyms anchor.

4 / 5

Distinctiveness Conflict Risk

It carves out a clear niche (the AWS MCP Server specifically) with distinct triggers like 'aws:ViaAWSMCPService and aws:CalledViaAWSMCP condition keys', making it highly unlikely to fire for the wrong skill.

5 / 5

Total

19

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

mreferre/standard-plugin-agent-toolkit-aws

Path

skills/aws-mcp-server/SKILL.md

Reviewed

Table of Contents

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.