My AI Guide
Measured conclusion: Deep-tested. L2 outcome: tools_ok.
Read-only access to the My AI Guide directory of AI tools, repos, plugins, news and guides.
| Author | co.myaiguide |
|---|---|
| Access | remote |
| Publisher key required | No |
| Measured state | Deep-tested |
| L1 liveness | alive |
| L1 latency | 2932 ms |
| Last measured | 2026-10-24 |
Add to your agent
Copy this into your MCP client config:
{
"mcpServers": {
"My AI Guide": {
"url": "https://myaiguide.co/api/mcp",
"transport": "streamable-http"
}
}
}
Measured tools (18)
searchread-onlyFull-text search across the AI tool directory and the repos/plugins catalog. Returns up to 10 matching tools and/or repos with their slugs and canonical URLs.
Parameters:
query(required) — string — Search text (min 2 characters).type— string — Which catalogs to search. Default: all.
Hints: read-only yes · destructive no · idempotent yes · open-world no
fetchread-onlyRetrieve the full text of a myaiguide.co document by id. The id is a canonical myaiguide.co URL or any myaiguide.co path (e.g. '/tools/claude' or 'https://myaiguide.co/about'). Returns id, title, text, and url.
Parameters:
id(required) — string — A myaiguide.co URL or site path, e.g. 'https://myaiguide.co/tools/claude' or '/about'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_site_sectionsread-onlyList every public section of myaiguide.co: tools, repos, blog, news, glossary, comparisons, categories, leaderboards, and the editorial/trust pages (about, disclosure, methodology, corrections, privacy, terms, etc.). Each section has a path, title, description and URL.
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_pageread-onlyFetch the readable text content of any page on myaiguide.co by its path (e.g. '/about', '/disclosure', '/methodology', '/tools/claude'). Reads editorial and trust pages and any other public page. Only paths on myaiguide.co are allowed.
Parameters:
path(required) — string — Site-relative path beginning with '/', e.g. '/about'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_toolsread-onlyList published AI tools, optionally filtered by category and pricing model, sorted by editor rating or recency. Paginated.
Parameters:
page— number — 1-based page number. Default 1.sort— string — Default: rating.category— string — Category slug, e.g. 'coding'.pageSize— number — Items per page (max 100). Default 24.pricingModel— string — e.g. 'free', 'freemium', 'paid'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_toolread-onlyFetch the full editorial record for one AI tool by slug: rating, pricing, pros/cons, target audience, use cases, FAQ, and the affiliate-safe /go redirect link.
Parameters:
slug(required) — string — Tool slug, e.g. 'claude-code'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_categoriesread-onlyList all tool categories with their slugs and descriptions.
Hints: read-only yes · destructive no · idempotent yes · open-world no
compare_toolsread-onlyFetch the editorial records for 2 to 3 tools side by side so you can compare them. Also returns the canonical /compare page URL.
Parameters:
slugs(required) — array — 2 to 3 tool slugs, e.g. ['claude','chatgpt'].
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_comparisonsread-onlyList published editorial comparison pages.
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_reposread-onlyList published GitHub repos, Claude Code plugins, MCP servers, and AI skills with install info, stars, and descriptions. Paginated.
Parameters:
page— integer — Page number, starting at 1 (default 1).limit— integer — Repos per page (default 25, max 100).
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_reporead-onlyFetch the full record for one repo/plugin by slug.
Parameters:
slug(required) — string — Repo or plugin slug, e.g. 'safishamsi-graphify'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
search_newsread-onlyRead the daily AI news roundups and weekly digests: new models, releases, features and pricing from the past days, newest first. Call it with no arguments to get the latest roundups. Each entry has a date, a summary and a link to its source.
Parameters:
type— string — Optional filter: 'daily-roundup' or 'weekly-digest'. Leave empty for both.limit— number — Max results (default 20, max 50).query— string — Optional search text. Leave empty to get the newest roundups.
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_tool_changelogread-onlyFetch recent changelog entries for one tool by slug.
Parameters:
slug(required) — string — Tool slug.limit— number — Max entries (default 20).
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_glossaryread-onlyList plain-English AI glossary terms with their definitions.
Parameters:
limit— number — Max terms (default 200).
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_glossary_termread-onlyFetch one glossary term by slug, with its extended explanation and FAQ.
Parameters:
slug(required) — string — Glossary term slug, e.g. 'mcp-model-context-protocol'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
list_blogread-onlyList published blog posts (guides and tutorials), newest first.
Parameters:
limit— number — Max posts (default 20).category— string — Optional blog category slug to filter by.
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_blog_postread-onlyFetch the full body of one blog post by slug.
Parameters:
slug(required) — string — Blog post slug, e.g. 'claude-code-guide'.
Hints: read-only yes · destructive no · idempotent yes · open-world no
get_site_contextread-onlyReturns publisher identity, company registration, editorial contact, the site's affiliate-link disclosure, its canonical URL patterns, and the locations of its machine-readable files (llms.txt, /for-ai, sitemap.xml).
Hints: read-only yes · destructive no · idempotent yes · open-world no
Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.