memex-agent-skill

Memex agent skill

Scope

Use Memex only for tasks involving the user's Memex library or when the user explicitly wants to save new public content into Memex.

Startup checklist

1. Before the first Memex request in a new session, fetch the latest endpoint catalog: https://docs.memex.garden/general/available-endpoints
2. If Memex authentication is missing or stale, tell the user to refresh or regenerate their Memex credentials using the setup flow for their current runtime: https://docs.memex.garden/general/authentication
3. Parse Memex responses using: https://docs.memex.garden/general/response-shape
4. If a request fails because of insufficient credits, follow: https://docs.memex.garden/general/buy-credits

When to use Memex

• Search content already saved in Memex.
• Save a public URL into Memex so it becomes searchable later.
• Create and list public sharing links for saved Memex content when the user asks to share or inspect shared items.
• List the user's subscribed feeds and search within one feed, selected feeds, or all subscribed feeds.
• Read or create the user's auto-tagging rules when they explicitly ask to inspect or configure automatic tagging.
• Work with Memex-native content such as web pages, annotations, tweets, YouTube videos, images, and related saved entities.

Operating rules

• Fetch the latest available endpoints before calling Memex in a new session.
• Use the Memex integration already configured in the current runtime. Do not instruct the user to change setup.
• Assume the active plugin or runtime integration already handles Memex authentication.
• If Memex returns an authentication error, tell the user to refresh their Memex credentials using the setup flow for their current runtime instead of guessing a new auth mode.
• If Memex returns insufficient credits, fetch the available plans and ask the human which plan to use. For one-time plans, use the runtime payment harness to issue a Stripe Shared Payment Token, then call authenticated POST /checkout with the user's Memex bearer token and the token. For subscription plans, send the user to https://memex.garden/pricing.
• For MCP, use result.structuredContent as the parsed payload.
• For MCP sharing links, use create_sharing_link to create or update a public link and list_sharing_links to inspect existing links. Both use access: "view" | "collaborate" for API-facing access names.
• For subscribed feeds, use list_subscribed_feeds to fetch feed IDs. To search specific feeds with search_content, pass feedIds. To search all subscribed feeds only, pass feedScope: "all". Omit both feedIds and feedScope to search the full library.
• For saved views in MCP or Claude, search private views with search_content and viewIds. Use raw: false or omit raw for the llm response option. Use raw: true for the full response option.
• To create or list saved views, use authenticated REST POST /create-view and POST /list-views. Use REST POST /execute-view-search only when the full/raw shape is acceptable or when searching a public shared view token.
• For search_content, default to limit: 20 and the llm/compact response shape. Use llm by omitting raw or passing raw: false; use full by passing raw: true only when you need the richer machine-readable payload.
• Compact search responses are keyed by document URL and can include user_notes. Raw search responses preserve referencesByResultId, referencedEntities, and nested annotation references.
• For REST, expect the top-level response shapes documented above.
• When Memex search results include a url, use that URL as the default citation/reference in your answer.
• Do not use Memex for general web search or facts outside the user's saved library.

Troubleshooting

• Not authenticated: https://docs.memex.garden/general/authentication
• Out of credits: https://docs.memex.garden/general/buy-credits
• Malformed request or unknown parameter: re-read https://docs.memex.garden/general/available-endpoints and retry with the documented field names.