[mcp-analysis] MCP Structural Analysis - 2026-09-09

[github-actions[bot]]

Tested 10 representative GitHub MCP tools (one per available toolset) plus the injected workflow-context block. Total payload today: 10,610 tokens, average usefulness 3.3/5. Best: github_context and list_workflows (5/5, wrapper-clean schemas, zero icon bloat). Worst: search_code (1/5 — hit a GitHub secondary rate limit, second time in 11 measured days). The persistent _meta.serverInfo icon block (~2.2k chars) continues to add fixed overhead to nearly every non-wrapper tool call, and one issue has been redacted by an integrity-policy filter on every single measured day (11/11).

Full Structural Analysis Report

Executive Summary

Metric	Value
Tools Analyzed	10
Total Tokens (Today)	10,610
Average Usefulness Rating	3.3/5
Best Rated Tool	github_context / list_workflows: 5/5
Worst Rated Tool	search_code: 1/5

Usefulness Ratings for Agentic Work

Tool	Toolset	Rating	Assessment
github_context (prompt block)	workflow_context	⭐⭐⭐⭐⭐	Zero-cost, immediately actionable identity metadata
list_workflows (mcpscripts wrapper)	actions	⭐⭐⭐⭐⭐	Clean JSON, explicit pagination, no icon bloat
list_code_scanning_alerts	code_security	⭐⭐⭐⭐	Rich remediation detail, but inconsistent schema across scanners
list_pull_requests	pull_requests	⭐⭐⭐⭐	Flat, minimal, no icon overhead
get_file_contents	repos	⭐⭐⭐	Content varies wildly by target file; fixed icon overhead
list_discussions	discussions	⭐⭐⭐	Good pagination metadata; icon block dominates small payload
search_users	users	⭐⭐⭐	Minimal identity fields only
get_label	labels	⭐⭐⭐	Tiny payload, icon overhead dominates
list_issues	issues	⭐⭐	Integrity-redacted item every day (11/11), blocks assessment
search_code	search	⭐	Hard 429 rate-limit error, zero usable data

Schema Analysis

Tool	Type	Depth	Key Fields	Notes
github_context	markdown	1	actor, repository, workspace, workflow-run-id	Prompt-injected, no MCP call
get_file_contents	object	3	content.resource.text, _meta.serverInfo	Base64 icon block ~2.2k chars fixed overhead
list_issues	array	2	number, title, state, labels	1 item redacted by integrity filter
list_pull_requests	array	1	number, title, state, html_url	Flat, no icon overhead
list_workflows (wrapper)	object	2	total_count, per_page, workflows[]	Best-in-class, camelCase perPage respected
list_code_scanning_alerts	array	3	rule.id, rule.severity, most_recent_instance.location	12 open critical/high alerts; third-party-tool entries have empty rule fields
list_discussions	object	3	discussions[].category.name, pageInfo, totalCount	Nested category/user relationships
get_label	object	1	name, color, description, id	Tiny payload, dominated by icon block
search_users	object	2	items[].login, items[].id, total_count	No name/bio/role fields
search_code	error	1	isError, rate-limit message	429 secondary rate limit

Response Size Analysis

Toolset	Avg Tokens (30d)	Tools Tested
repos	2,176	1
code_security	1,979	1
labels	956	1
discussions	685	1
search	616	1
users	612	1
actions	171	1
workflow_context	94	1
issues	61	1
pull_requests	38	1

Tool-by-Tool Analysis (Today)

Tool	Toolset	Tokens	Schema	Rating	Notes
github_context	workflow_context	90	markdown	⭐⭐⭐⭐⭐	Zero MCP cost
list_workflows	actions	175	object	⭐⭐⭐⭐⭐	mcpscripts wrapper, no bloat
list_code_scanning_alerts	code_security	6,900	array	⭐⭐⭐⭐	12 alerts, mixed scanner schemas
list_pull_requests	pull_requests	35	array	⭐⭐⭐⭐	Minimal, actionable
get_file_contents	repos	780	object	⭐⭐⭐	.gitignore target, icon overhead
list_discussions	discussions	700	object	⭐⭐⭐	Icon overhead dominates
search_users	users	640	object	⭐⭐⭐	Needs follow-up call for detail
get_label	labels	610	object	⭐⭐⭐	Icon overhead dominates
list_issues	issues	60	array	⭐⭐	Integrity-redacted item
search_code	search	620	error	⭐	429 rate limit, no data

30-Day Trend Summary

Metric	Value
Data Points	110
Date Range	2026-08-25 to 2026-09-09 (11 measured days)
Average Daily Total Tokens	7,388
Average Rating Trend	Stable (~3.3–3.7/5 across all measured days)

Recommendations

• High-value tools (rating 4-5): github_context, list_workflows (mcpscripts wrapper), list_pull_requests, list_code_scanning_alerts — prefer these first when building agentic workflows against this repo.
• Tools needing improvement: list_issues (integrity-redacted item blocks visibility every day, 11/11 measured), search_code (recurring secondary rate-limit failures, now 2/11 days).
• Context-efficient tools (low tokens, high rating): github_context (90 tokens, zero MCP cost), list_workflows wrapper (175 tokens).
• Context-heavy tools: get_file_contents on repos (up to 7,500 tokens depending on target file) and list_code_scanning_alerts (up to 6,900 tokens on a busy day) — use field/pagination limits and prefer directory listings over full-file reads where possible.

Visualizations

Response Size by Toolset

Usefulness Ratings

Daily Token Trend

Size vs Usefulness

References: §34347987159

Warning

Firewall blocked 1 domain

The following domain was blocked by the firewall during workflow execution:

• api.anthropic.com

To allow these domains, add them to the network.allowed list in your workflow frontmatter:

network:
  allowed:
    - defaults
    - "api.anthropic.com"

See Network Configuration for more information.

Generated by 🔍 GitHub MCP Structural Analysis · claude · agent · 164.3 AIC · ⌖ 10.2 AIC · ⊞ 11.5K · ◷

• expires on Sep 10, 2026, 4:03 AM UTC-08:00

[github-actions[bot]]

This discussion has been marked as outdated by GitHub MCP Structural Analysis.

A newer discussion is available at Discussion #59963.