01Safety
0.0 / 30
What changed in the harness
Selection accuracy 98→98, token cost up 8%, unconfirmed writes 0%→0%.
Category breakdown
Where the score comes from.
Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.
02Legibility
26.9 / 30
03Economics
19.0 / 20
04Discoverability
17.7 / 20
Highest-impact fix
Estimated gain +30 pointsAdd explicit identity and permission preflight tools
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions.
Description evidence
Defects and rewrites.
8 defects found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.
| Tool | Defect types | Suggested rewrite |
|---|---|---|
ask |
no_return_description |
Execute a task or question through GitHub Copilot CLI, supporting @ file/image references (e.g., @file.png, @src/) and a ! prefix for direct shell commands, with control over model, file access, tools, and safety settings. Returns the response and output generated by Copilot for the prompt. |
batch |
no_return_description |
Delegate multiple atomic tasks to GitHub Copilot CLI for batch processing, ideal for repetitive operations, mass refactoring, and automated code transformations, with options to run tasks in parallel, stop on error, and control model and tool access. Returns the results and outputs produced for each delegated task. |
review |
no_return_description |
Perform a comprehensive code review of a target using GitHub Copilot CLI, supporting multiple review types (security, performance, quality, architecture, testing, accessibility, etc.) and configurable severity filtering, output formats, fix suggestions, and issue limits. Returns the review report with findings and prioritized issues. |
ping |
no_return_description |
Send a test message and receive it echoed back, used to verify that the server is reachable and responsive. Returns the echoed message. |
Help |
name_restates_behavior no_return_description |
Show usage details, available commands, and configuration guidance for the Copilot CLI when you need help getting started or resolving setup questions. Returns the help information text for the Copilot CLI. |
version |
name_restates_behavior no_return_description |
Retrieve the installed versions and system information for GitHub Copilot and this MCP server, useful for diagnostics and compatibility checks. Returns the version and system details as text. |
brainstorm |
no_return_description |
Generate creative ideas for a brainstorming challenge using structured frameworks (divergent, convergent, scamper, design-thinking, lateral), optionally configured with a domain, constraints, existing context, and idea count. Returns the generated ideas, optionally with feasibility and impact analysis. |
timeout-test |
name_restates_behavior no_return_description |
Verify that the server prevents timeout failures by running for a specified duration and completing the call successfully. Returns a confirmation of successful completion after the requested duration elapses. |
Selection evidence
Confusable tool pairs.
0 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
| No confusable tool pairs were flagged in this assessment. | |||
Compare the field