npx @xmemo/client setup --url https://xmemo.dev<\/code> to auto-generate per-client instance IDs and apply reviewed config for supported clients.", "mcpSetupConsole": "MCP setup console", "mcpSetupDescription": "Pick the MCP client to see the public-safe snippet for its settings surface. OAuth snippets omit bearer tokens; direct snippets reference XMEMO_KEY<\/code> only when needed plus generated, non-secret agent identity headers instead of literal credentials.", "canonicalCapabilityMatrix": "Open the canonical client capability matrix", "mcpDocsRedirect": "MCP setup details, client snippets, and tool explanations are organized in Docs.", "navTrust": "Trust", "noMatches": "No matches.", "openAccountEntry": "Open account entry ↗", "openDocsChapter": "Open Docs chapter", "openProductNavigation": "Open public navigation", "pasteSnippet": "Paste the selected snippet into the listed config path or dashboard, keep the hosted URL as shown, and set XMEMO_AGENT_INSTANCE_ID once per local client when the snippet references it. Then complete OAuth for OAuth clients or restart the direct MCP client.", "personaConnection": "Connection", "personaFlowsBody": "Choose the path that matches how you want to use XMemo. Each flow keeps credentials out of public docs and ends with a concrete memory action.", "personaFlowsTitle": "Persona onboarding flows", "personaGoal": "Goal", "personaSuccessAction": "Success action", "personaTroubleshooting": "Troubleshooting", "perClientMcpConfigPreview": "Per-client MCP config preview", "pilotAccess": "Pilot access", "product": "Home", "productDesc": "Back to the XMemo overview.", "productHome": "Home", "publicNavigation": "XMemo public navigation", "publicResources": "XMemo public resources", "loadingProductContext": "Loading product context", "productContextUnavailable": "Product context unavailable", "cachedPublicResourceRoute": "Showing the cached public resource route while the context API recovers.", "publicSafeNote": "Public-safe page: only hosted public URLs are shown; no console links, credential values, or internal control-plane endpoints are exposed here.", "quickSetup": "Quick setup", "quickstart": "Quickstart for XMemo", "search": "Search", "searchDocs": "Search docs", "searchPlaceholder": "Search docs, MCP, skills, install…", "stories": "Stories", "storiesDesc": "Milestones, ecosystem launches, and source-backed XMemo updates.", "skills": "Skills", "skillsDesc": "Connect XMemo as an agent Skill in addition to MCP.", "skillsDocsRedirect": "Skill connector guidance and agent command patterns are organized in Docs.", "start": "Get started", "startPilot": "Start pilot", "startPilotDesc": "Jump to the public pilot path.", "supportedMcpClients": "Supported MCP clients", "typeToFilter": "Type to filter", "brandHome": "XMemo home", "language": "Language", "changelog": "Changelog", "downloadSkillPackage": "Download Skill package", "installSkillTitle": "Install the Skill", "installTool": "Download tool", "curlInstall": "curl", "wgetInstall": "wget", "macosLinuxInstall": "macOS, Linux, and WSL", "windowsInstall": "Windows PowerShell", "askAgentInstall": "Ask your agent", "inspectInstallScript": "Inspect install script", "wgetAlternative": "Alternative with wget", "posixInstallExplanation": "Use this in macOS, Linux, or WSL.", "powershellInstallExplanation": "Use this in Windows PowerShell.", "agentInstallExplanation": "Paste this into an agent you trust to follow its normal safety checks.", "copyInstallCommand": "Copy install command", "posixInstallCommand": "macOS, Linux, and WSL", "powershellInstallCommand": "Windows PowerShell", "copyPowerShellInstallCommand": "Copy PowerShell install command", "copyFailed": "Copy failed. Select the command text, then press Ctrl+C or Command+C."}, "docsMarkdown": ["# Quickstart for XMemo", "", "Hosted XMemo quickstart for connecting MCP clients, CLIs, IDEs, and autonomous agents without putting bearer tokens in public configuration.", "", "- CLI package: @xmemo/client", "- Service URL: https://xmemo.dev", "- MCP URL: https://xmemo.dev/mcp", "- Setup command: npx @xmemo/client setup --url https://xmemo.dev", "- Direct MCP token environment variable: XMEMO_KEY", "- OAuth-supported clients: VS Code / GitHub Copilot, ChatGPT / OpenAI Apps SDK, Gemini CLI, Cursor, Autonomous agent", "- Persona flows: ChatGPT user, Copilot/Codex developer, team pilot owner, autonomous agent operator", "- Autonomous agent preset: OAuth-first when supported; XMEMO_KEY fallback for headless runners; runner-generated XMEMO_AGENT_INSTANCE_ID required", "- Supported MCP clients: vscode, chatgpt, claude, copilot-cli, codex, gemini-cli, cursor, autonomous-agent, other", "- MCP chapter: per-client setup snippets and hosted streamable-http boundary", "- Skills chapter: standalone Skill runtime, device login, and direct HTTPS operations", "- SDK helpers: taskStart, taskEnd, recordDecision, recordBugFix, recordCorrection, captureEvent (recordEvent is an alias)", "- Retrieval tool references: recall (https://xmemo.dev/docs/tools/recall), search_memory (https://xmemo.dev/docs/tools/search), recall_context (https://xmemo.dev/docs/tools/recall-context)", "Choosing the right retrieval tool: recall is a lightweight, forgiving lookup; search_memory is a strict, targeted search; recall_context packages multiple bounded memories into task context.", "- Troubleshooting: docs/troubleshooting (https://xmemo.dev/docs/troubleshooting)", "- Safety guarantees: manual config review, environment-variable secrets, no token in generated URLs", "", "## Hosted XMemo quickstart", "", "Start with the managed XMemo service and one of these connection paths. Self-hosted, Docker, Supabase, and Python package setup are developer references, not the first path for hosted users.", "", "- OAuth clients: https://xmemo.dev/mcp (memory:read memory:write)", "- @xmemo/client setup: npx @xmemo/client setup --url https://xmemo.dev", "- Direct MCP with XMEMO_KEY: https://xmemo.dev/mcp", "", "## Memory Console controls", "", "XMemo is not only an MCP endpoint. The Memory Console is the user-facing place to review, correct, remove, and export memory while keeping agent attribution visible.", "", "- View and search memory", "- Edit or correct entries", "- Delete and export", "- Agent attribution", "- Privacy and credential controls", "", "## FAQ / Troubleshooting", "", "Use these answers when a user or marketplace reviewer gets blocked. Every answer includes an action step and a safety boundary."], "skillPackage": null, "skillCapability": null, "docsNavTree": null, "activeRoute": null, "activeToolReference": null}, "prefixedRoutesActive": false}
{"schemaVersion": "1.0.0", "pageId": "product-stories", "locale": "en-US", "contentLocale": "en-US", "status": "native", "fallbackReason": "", "canonicalPath": "/product/stories", "routePath": "/product/stories", "family": "product_subpage", "title": "Customer Stories & Workflows", "description": "", "editionDigest": "b63e172bf2706a5d9cb2263b414117ff98e6e2f8039abf4d70c84b8ffaa50829", "revision": "b63e172bf2706a5d9cb2263b414117ff98e6e2f8039abf4d70c84b8ffaa50829", "canonicalContentDigest": "59fd49ada3627de3cf3ec125831d792bbfa3f10de3537b3c084a9285d04215f5", "anchors": ["stories-catalog"], "serverOwned": false, "metadata": {"title": "Customer Stories & Workflows", "description": "", "canonicalPath": "/product/stories", "htmlLang": "en"}, "payload": {"page": "stories", "content": {"label": "Stories", "title": "Stories", "kicker": "Progress and proof", "lead": "Milestones, ecosystem launches, and source-backed adoption signals behind XMemo.", "cards": [], "sections": []}, "siblings": [{"slug": "docs", "label": "Docs", "active": false}, {"slug": "mcp", "label": "MCP", "active": false}, {"slug": "skills", "label": "Skills", "active": false}, {"slug": "changelog", "label": "Changelog", "active": false}, {"slug": "stories", "label": "Stories", "active": true}], "cards": [], "sections": [], "mcpClientSnippets": [], "stories": [{"slug": "chatgpt-project-memory-workspace", "type": "launch", "status": "published", "title": "Your project memory, visible inside ChatGPT", "summary": "XMemo Project Workspace turns a ChatGPT request into a visual, navigable memory workspace inside the conversation, so durable project context is easier to inspect, use, and control.", "body": "## Stop re-explaining the project\n\nLong-running work rarely fits in one chat. Decisions move across conversations, useful context gets buried, and the next session often begins with the same briefing all over again. XMemo already gives ChatGPT durable memory across conversations. Project Workspace makes that memory visible where the work is happening.\n\nAsk ChatGPT to open the project workspace and XMemo returns an interactive panel inside the conversation. It is not a separate dashboard or a static summary. It is a focused view of the authorized project records that ChatGPT can use with you.\n\n## One workspace, eight ways into the project\n\nThe workspace organizes project context into eight visible destinations: Overview, Memory, TODO, Decisions, Documents, Conversations, Timeline, and Settings. Each view answers a different question without making the user reconstruct the project from raw memory search results.\n\n- Overview summarizes the current project and its recent state.\n- Memory surfaces the durable facts, preferences, constraints, and working context linked to the project.\n- TODO and Decisions separate future work from choices that already shape the project.\n- Documents and Conversations preserve the sources behind the work.\n- Timeline puts changes in sequence, while Settings keeps project controls close at hand.\n\n## Memory you can actually inspect\n\nThe Memory view is designed for clarity before volume. Each row shows a title, memory type, review state, source, and update date. Selecting a row opens its detail in the same panel, where an authorized user can inspect the full content and edit it without leaving ChatGPT.\n\nResults arrive in bounded pages of ten records with an explicit Load more action. That keeps the first view fast and readable while preserving a deliberate path into deeper project history.\n\nThe screenshot above comes from the current XMemo Project Workspace React implementation using synthetic project data. It shows the real Memory list and selected-memory detail layout, not a concept mockup.\n\n## Built around user control\n\nA memory interface should explain what it is showing, not turn remembered context into an invisible side effect. Project Workspace keeps provenance, status, review state, and modification controls in view. The server still applies authorization before records reach the widget, and sensitive host capabilities remain private to the active ChatGPT session rather than appearing in rendered content.\n\nThis is the practical shape of user-owned AI memory: ChatGPT can recall the project, the user can see what was recalled, and both can work from the same durable context without surrendering control of it.\n\n## Source note\n\nThis is a self-reported XMemo product update describing the Project Workspace implementation available in the XMemo ChatGPT app codebase as of July 30, 2026. Feature availability depends on the connected XMemo app, the user's authorization, and the project data available to that account.", "date": "2026-07-30", "audience": "users", "sourceUrl": "https://xmemo.dev/product/mcp", "sourceLabel": "XMemo MCP and ChatGPT setup", "proofLevel": "self-reported", "tags": ["chatgpt", "project workspace", "memory", "visual workspace", "mcp apps"], "featured": false, "heroImage": "/assets/stories/chatgpt-project-memory-workspace.png", "heroImageAlt": "XMemo Project Workspace open inside ChatGPT, showing a project memory list and the selected memory detail panel.", "heroImageCaption": "The live Project Workspace component rendered with synthetic project data. The Memory view keeps provenance, review state, and editing controls visible inside ChatGPT.", "ogImage": "https://xmemo.dev/assets/stories/chatgpt-project-memory-workspace.png", "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/chatgpt-project-memory-workspace", "plainMarkdownUrl": "https://xmemo.dev/product/stories/chatgpt-project-memory-workspace.md", "dateModified": "2026-07-30", "schemaType": "BlogPosting"}, {"slug": "agent-self-registration", "type": "launch", "status": "published", "title": "AI agents can start with XMemo without signing up", "summary": "A compatible AI agent can use isolated temporary memory before a user connects an account. The user decides later what to keep.", "body": "## Start without an account\n\nXMemo now supports a no-sign-up start for compatible AI agents. When a deployment operator enables it, an agent can register without a formal user token. XMemo returns a temporary token, a claim code, and a link for connecting the agent to a user later.\n\n## Temporary memory only\n\nThe temporary token works only inside that agent's isolated memory space. It can store and recall temporary notes, but it cannot open a user's existing memories, workspace data, account settings, or another agent's memory.\n\n## Connect it when you are ready\n\nA user can open the link and claim the agent at any time. During that step, the user chooses which temporary notes to keep and which to discard. Nothing is added to the user's permanent memory without that choice.\n\n## This is not anonymous access\n\nThe feature is off by default and must be enabled by a deployment operator. Temporary memory has size and time limits, the token expires, and registration is rate-limited. Signing up normally remains the default path when a user is available.\n\n## How to verify it\n\nThe public XMemo agent manifest lists the registration endpoint, temporary-token scopes, available memory tools, isolation boundary, and claim flow. XMemo's regression coverage also checks token expiry, sandbox isolation, rate limits, and the keep-or-discard decision.", "date": "2026-07-16", "audience": "developers", "sourceUrl": "https://xmemo.dev/.well-known/xmemo-agent.json", "sourceLabel": "XMemo agent setup manifest", "proofLevel": "self-reported", "tags": ["agent onboarding", "no sign-up", "temporary memory", "claim flow", "privacy"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/agent-self-registration", "plainMarkdownUrl": "https://xmemo.dev/product/stories/agent-self-registration.md", "ogImage": "https://xmemo.dev/product/stories/agent-self-registration/og.svg", "dateModified": "2026-07-16", "schemaType": "BlogPosting"}, {"slug": "own-database-memory-storage", "type": "launch", "status": "published", "title": "XMemo adds Own Database storage for AI memory", "summary": "Teams can keep XMemo's hosted recall layer while routing memory data to their own PostgreSQL or Supabase database.", "body": "## What happened\n\nXMemo added Own Database storage for AI memory. Users can register a PostgreSQL-compatible database, including Supabase, verify the connection, initialize the memory schema through an explicit setup flow, and choose whether new memories are written to XMemo Cloud or to their own database.\n\n## Why it matters\n\nAI memory is long-lived project context: decisions, preferences, handoffs, and agent activity that should remain portable and governable. Own Database gives teams more control over where that memory data lives while preserving XMemo's hosted recall, search, and multi-agent coordination experience.\n\n## How it works\n\nThe hosted XMemo control plane still manages accounts, sessions, policies, setup state, and redacted diagnostics. The memory data plane can route to XMemo Cloud or to the user's selected PostgreSQL or Supabase backend after verification and activation.\n\n## Setup and sync boundaries\n\nDatabase setup is explicit: users review connection status, dry-run initialization, apply migrations, and activate the backend before new writes move to Own Database. Existing memories are not silently moved. XMemo provides manual Cloud-to-database and database-to-Cloud sync previews with conflict review before applying changes.\n\n## Source note\n\nThis is a self-reported XMemo product update based on the implemented Storage settings, backend routing, Supabase setup, and manual sync flows. Own Database is a storage ownership feature; it is not a zero-knowledge, end-to-end encryption, or silent automatic migration claim.", "date": "2026-06-29", "audience": "enterprise", "sourceUrl": "https://xmemo.dev/product/stories", "sourceLabel": "XMemo Stories", "proofLevel": "self-reported", "tags": ["own database", "byodb", "postgresql", "supabase", "storage ownership"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/own-database-memory-storage", "plainMarkdownUrl": "https://xmemo.dev/product/stories/own-database-memory-storage.md", "ogImage": "https://xmemo.dev/product/stories/own-database-memory-storage/og.svg", "dateModified": "2026-06-29", "schemaType": "BlogPosting"}, {"slug": "hermes-xmemo-plugin-atlas", "type": "marketplace", "status": "published", "title": "Hermes Atlas lists the XMemo memory provider plugin", "summary": "Hermes users can discover `hermes-xmemo`, a native memory provider that connects Hermes Agent to XMemo's durable, user-owned memory layer.", "body": "## What happened\n\nHermes Atlas lists `yonro/hermes-xmemo-plugin` as an XMemo memory provider for Hermes Agent. The package name is `hermes-xmemo`, and the Hermes plugin name is `xmemo`.\n\n## Why it matters\n\nHermes Agent can use XMemo as a native memory provider instead of treating XMemo only as a generic MCP endpoint. The plugin participates in the Hermes memory lifecycle with pre-turn recall, native memory mirroring, working state, optional workflow tools, and session-end snapshots.\n\n## What it provides\n\nThe plugin connects Hermes to a user-owned XMemo account for durable cross-session memory, semantic search, cross-agent recall, state storage with TTL, and background prefetching. XMemo's hosted MCP endpoint remains available as a portable fallback for clients that do not have a native Hermes integration.\n\n## Evidence\n\nThe source link points to the public Hermes Atlas listing. The listing describes the plugin as a native memory provider for durable cross-session AI memory with semantic search and links back to the `yonro/hermes-xmemo-plugin` repository.", "date": "2026-06-26", "audience": "developers", "sourceUrl": "https://hermesatlas.com/projects/yonro/hermes-xmemo-plugin", "sourceLabel": "Hermes Atlas listing", "proofLevel": "public-source", "tags": ["hermes", "hermes atlas", "plugin", "memory provider", "semantic memory"], "featured": false, "locale": "en-US", "sourceDomain": "hermesatlas.com", "canonicalUrl": "https://xmemo.dev/product/stories/hermes-xmemo-plugin-atlas", "plainMarkdownUrl": "https://xmemo.dev/product/stories/hermes-xmemo-plugin-atlas.md", "ogImage": "https://xmemo.dev/product/stories/hermes-xmemo-plugin-atlas/og.svg", "dateModified": "2026-06-26", "schemaType": "BlogPosting"}, {"slug": "xmemo-openclaw-plugin-clawhub", "type": "marketplace", "status": "published", "title": "XMemo for OpenClaw is available on ClawHub", "summary": "OpenClaw users can install the XMemo plugin from ClawHub as a native memory provider for long-term memory, semantic recall, TODOs, snapshots, and audit tools.", "body": "## What happened\n\nXMemo for OpenClaw became available on ClawHub on June 19, 2026. The package `@xmemo/openclaw-memory` gives OpenClaw a native `kind: \"memory\"` provider for XMemo Cloud Memory.\n\n## Why it matters\n\nThe plugin is the runtime integration path. Once enabled, OpenClaw can use XMemo as its active long-term memory backend instead of relying only on local file-backed or vector-backed stores.\n\n## What it provides\n\nThe plugin exposes XMemo memory reads and writes, semantic recall, memory update and forget flows, TODO tools, timeline events, restart snapshots, and audit-oriented tools. It uses the shared XMemo credential model and can pair with the XMemo Memory Skill for recall-first workflow guidance.\n\n## Evidence\n\nThe source link points to the public ClawHub plugin listing. This story covers the native plugin distribution path; the XMemo Memory Skill remains a separate ClawHub entry for agent instructions and workflow guidance.", "date": "2026-06-19", "audience": "developers", "sourceUrl": "https://clawhub.ai/plugins/@xmemo/openclaw-memory", "sourceLabel": "ClawHub plugin listing", "proofLevel": "official", "tags": ["openclaw", "clawhub", "plugin", "mcp", "memory"], "featured": false, "locale": "en-US", "sourceDomain": "clawhub.ai", "canonicalUrl": "https://xmemo.dev/product/stories/xmemo-openclaw-plugin-clawhub", "plainMarkdownUrl": "https://xmemo.dev/product/stories/xmemo-openclaw-plugin-clawhub.md", "ogImage": "https://xmemo.dev/product/stories/xmemo-openclaw-plugin-clawhub/og.svg", "dateModified": "2026-06-19", "schemaType": "BlogPosting"}, {"slug": "xmemo-listed-on-smithery", "type": "marketplace", "status": "published", "title": "XMemo is listed on Smithery", "summary": "XMemo's hosted MCP server is easier to discover through Smithery, giving MCP users a clearer path to connect a user-owned memory layer.", "body": "## What happened\n\nXMemo now has a public Smithery server listing for its hosted MCP memory layer.\n\n## Why it matters\n\nMCP users can find the server from a recognized MCP directory instead of relying only on private setup notes. The listing supports discovery, but it does not imply every marketplace or host has certified XMemo.\n\n## Evidence\n\nThe source link points to the public Smithery listing. Use this as an official directory signal, not as a broad partnership claim.", "date": "2026-06-13", "audience": "developers", "sourceUrl": "https://smithery.ai/servers/xmemo/xmemo", "sourceLabel": "Smithery listing", "proofLevel": "official", "tags": ["mcp", "marketplace", "discovery"], "featured": true, "featuredReason": "A public third-party MCP directory listing is a strong discovery signal for developers evaluating XMemo.", "locale": "en-US", "sourceDomain": "smithery.ai", "canonicalUrl": "https://xmemo.dev/product/stories/xmemo-listed-on-smithery", "plainMarkdownUrl": "https://xmemo.dev/product/stories/xmemo-listed-on-smithery.md", "ogImage": "https://xmemo.dev/product/stories/xmemo-listed-on-smithery/og.svg", "dateModified": "2026-06-13", "schemaType": "BlogPosting"}, {"slug": "public-product-transparency", "type": "trust", "status": "published", "title": "XMemo rebuilds public updates for clearer product transparency", "summary": "XMemo is separating changelog details, trust material, and progress stories so users can see what changed and why it matters.", "body": "## What happened\n\nXMemo is moving public product communication into clearer lanes: Docs for setup, Changelog for release notes, Trust for safety boundaries, and Stories for source-backed milestones.\n\n## Why it matters\n\nA memory product needs careful claims. Separating these pages helps users, teams, and investors distinguish implemented capability from roadmap or internal engineering notes.\n\n## Source note\n\nThis is a self-reported XMemo product update. This Stories section is the first version of that public content system. Claims are tagged by proof level and point to source links when available.", "date": "2026-06-13", "audience": "investors", "sourceUrl": "https://xmemo.dev/product/stories", "sourceLabel": "Stories index", "proofLevel": "self-reported", "tags": ["transparency", "trust", "product"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/public-product-transparency", "plainMarkdownUrl": "https://xmemo.dev/product/stories/public-product-transparency.md", "ogImage": "https://xmemo.dev/product/stories/public-product-transparency/og.svg", "dateModified": "2026-06-13", "schemaType": "BlogPosting"}, {"slug": "oauth-first-mcp-connections", "type": "launch", "status": "published", "title": "XMemo adds OAuth-first MCP connections", "summary": "XMemo's public setup guidance now separates OAuth-capable clients from direct-token clients, reducing raw credential handling during agent setup.", "body": "## What happened\n\nXMemo now documents OAuth-first setup for supported MCP clients and keeps direct XMEMO_KEY usage limited to clients that cannot complete OAuth.\n\n## Why it matters\n\nUsers and reviewers get a clearer security boundary: hosted OAuth where the client supports it, local secret stores for direct clients, and no bearer tokens in public examples.\n\n## Source note\n\nThis is a self-reported XMemo product update. The public MCP and Docs pages describe the supported connection lanes and the review-safe setup contract.", "date": "2026-06-12", "audience": "developers", "sourceUrl": "https://xmemo.dev/product/mcp", "sourceLabel": "XMemo MCP page", "proofLevel": "self-reported", "tags": ["oauth", "mcp", "security"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/oauth-first-mcp-connections", "plainMarkdownUrl": "https://xmemo.dev/product/stories/oauth-first-mcp-connections.md", "ogImage": "https://xmemo.dev/product/stories/oauth-first-mcp-connections/og.svg", "dateModified": "2026-06-12", "schemaType": "BlogPosting"}, {"slug": "agent-client-support-expands", "type": "community", "status": "published", "title": "XMemo expands setup guidance for agent clients", "summary": "The public docs now organize setup patterns for ChatGPT, GitHub Copilot, Codex, Claude, Gemini CLI, Cursor, and generic MCP clients.", "body": "## What happened\n\nXMemo's public Docs page now groups agent-client setup patterns by OAuth, direct MCP, and autonomous-runner flows.\n\n## Why it matters\n\nPeople use more than one AI client. A memory layer is more useful when setup guidance acknowledges that reality while keeping each client's credential model separate.\n\n## Source note\n\nThis is a self-reported XMemo product update. The public Docs page lists the supported profiles and describes how credentials should be handled for each lane.", "date": "2026-06-11", "audience": "users", "sourceUrl": "https://xmemo.dev/product/docs", "sourceLabel": "XMemo Docs", "proofLevel": "self-reported", "tags": ["agents", "docs", "setup"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/agent-client-support-expands", "plainMarkdownUrl": "https://xmemo.dev/product/stories/agent-client-support-expands.md", "ogImage": "https://xmemo.dev/product/stories/agent-client-support-expands/og.svg", "dateModified": "2026-06-11", "schemaType": "BlogPosting"}, {"slug": "persistent-agent-memory-workflow", "type": "case_study", "status": "published", "title": "How an AI coding workflow keeps memory across agents", "summary": "A source-backed workflow story shows how durable decisions and handoff notes can move between coding agents without exposing private memory content publicly.", "body": "## Context\n\nModern coding workflows often move between IDE assistants, CLI agents, and review tools. Without a shared memory layer, project decisions can vanish between runs.\n\n## Solution\n\nXMemo records scoped decisions, bug-fix notes, and task handoffs with agent attribution. Later agents can recall the approved context before editing.\n\n## Source note\n\nThis is an internal workflow story, not a named customer case study. It is published as self-reported product usage and avoids private memory content.", "date": "2026-06-10", "audience": "developers", "sourceUrl": "https://xmemo.dev/product/docs", "sourceLabel": "Workflow helpers in Docs", "proofLevel": "self-reported", "tags": ["workflow", "agents", "memory"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/persistent-agent-memory-workflow", "plainMarkdownUrl": "https://xmemo.dev/product/stories/persistent-agent-memory-workflow.md", "ogImage": "https://xmemo.dev/product/stories/persistent-agent-memory-workflow/og.svg", "dateModified": "2026-06-10", "schemaType": "BlogPosting"}, {"slug": "xmemo-chatgpt-apps-directory", "type": "marketplace", "status": "published", "title": "XMemo is available in the ChatGPT Apps Directory", "summary": "ChatGPT users can find XMemo, start a chat, and connect through the Apps Directory OAuth flow without pasting API keys or JSON configuration.", "body": "## What happened\n\nXMemo became available in the ChatGPT Apps Directory on June 10, 2026. The listed app uses XMemo's hosted memory layer so ChatGPT users can save and recall durable context across conversations.\n\n## Why it matters\n\nChatGPT is one of the primary places where people build long-running workflows, but useful context often stays trapped in a single conversation. XMemo gives ChatGPT users a dedicated memory layer for preferences, project context, guidelines, to-dos, and ledger-style notes that can be recalled later.\n\n## Connection model\n\nUsers connect from ChatGPT through the app listing and approve XMemo OAuth. The public path does not require users to paste raw API keys, bearer tokens, or JSON MCP configuration into ChatGPT.\n\n## Source note\n\nThis is a self-reported XMemo marketplace update based on the ChatGPT Apps approval and listing state observed on June 10, 2026. It should be read as an availability milestone for the ChatGPT Apps Directory, not as a broader certification claim for every OpenAI or MCP surface.", "date": "2026-06-10", "audience": "users", "sourceUrl": "https://xmemo.dev/product/mcp", "sourceLabel": "XMemo MCP and ChatGPT setup", "proofLevel": "self-reported", "tags": ["chatgpt", "apps directory", "oauth", "memory", "marketplace"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/xmemo-chatgpt-apps-directory", "plainMarkdownUrl": "https://xmemo.dev/product/stories/xmemo-chatgpt-apps-directory.md", "ogImage": "https://xmemo.dev/product/stories/xmemo-chatgpt-apps-directory/og.svg", "dateModified": "2026-06-10", "schemaType": "BlogPosting"}, {"slug": "security-mode-controls", "type": "trust", "status": "published", "title": "XMemo documents Cloud, Keyguard, and Vault memory boundaries", "summary": "A public security guide explains where memory data lives, who can decrypt it, and how users can verify the active security mode.", "body": "## What happened\n\nXMemo added public guidance for Cloud, Keyguard, and Vault security modes.\n\n## Why it matters\n\nUsers should know whether hosted recall can decrypt memory on their behalf, whether keys are customer-managed, or whether data is stored as zero-knowledge ciphertext.\n\n## Source note\n\nThis is a self-reported XMemo product update. The source guide describes storage, decryption, deletion, DSAR export, and verification boundaries for each mode.", "date": "2026-06-09", "audience": "enterprise", "sourceUrl": "https://xmemo.dev/docs/security/where-is-my-data", "sourceLabel": "Where is my data?", "proofLevel": "self-reported", "tags": ["security", "privacy", "trust"], "featured": false, "locale": "en-US", "sourceDomain": "xmemo.dev", "canonicalUrl": "https://xmemo.dev/product/stories/security-mode-controls", "plainMarkdownUrl": "https://xmemo.dev/product/stories/security-mode-controls.md", "ogImage": "https://xmemo.dev/product/stories/security-mode-controls/og.svg", "dateModified": "2026-06-09", "schemaType": "BlogPosting"}, {"slug": "xmemo-clawhub-skill", "type": "marketplace", "status": "published", "title": "XMemo Memory Skill is available on ClawHub", "summary": "OpenClaw users can install the XMemo Memory Skill from ClawHub so agents know when to recall, remember, preserve handoff state, and avoid saving secrets.", "body": "## What happened\n\nXMemo published its ClawHub Skill on June 9, 2026. OpenClaw users can install it with the ClawHub skill command and give agents a guided workflow for durable, user-owned memory across sessions, projects, and tools.\n\n## Why it matters\n\nSkills are the workflow layer for agents. The XMemo Memory Skill teaches agents when to recall context, when to save durable decisions or TODOs, how to preserve handoff state, and when not to store sensitive data.\n\n## Runtime boundary\n\nThe Skill is guidance, not the memory runtime. XMemo setup and diagnostics should use the `xmemo` CLI, while OpenClaw should use the native XMemo memory plugin for actual memory reads and writes. Generic clients can use XMemo's hosted MCP endpoint when they do not have a native integration.\n\n## Evidence\n\nThe source link points to the public ClawHub Skill listing. This story covers the Skill distribution path; the native OpenClaw plugin is tracked as a separate ClawHub marketplace entry.", "date": "2026-06-09", "audience": "developers", "sourceUrl": "https://clawhub.ai/xmemo/xmemo", "sourceLabel": "ClawHub Skill listing", "proofLevel": "official", "tags": ["openclaw", "clawhub", "skill", "agents", "memory"], "featured": false, "locale": "en-US", "sourceDomain": "clawhub.ai", "canonicalUrl": "https://xmemo.dev/product/stories/xmemo-clawhub-skill", "plainMarkdownUrl": "https://xmemo.dev/product/stories/xmemo-clawhub-skill.md", "ogImage": "https://xmemo.dev/product/stories/xmemo-clawhub-skill/og.svg", "dateModified": "2026-06-09", "schemaType": "BlogPosting"}], "storiesMeta": {"allStoriesLabel": "All stories", "filterLabel": "Filter stories", "allTypesLabel": "All types", "proofLevelsLabel": "Proof levels", "evidenceLabel": "Evidence", "sourceLabel": "Source", "readStoryLabel": "Read story", "viewSourceLabel": "View source", "backToStoriesLabel": "Back to Stories", "backToAllStoriesLabel": "Back to all Stories", "previousStoryLabel": "Previous story", "nextStoryLabel": "Next story", "relatedStoriesLabel": "Related stories", "noStoryFoundLabel": "Story not found", "typeLabels": {"award": "Award", "marketplace": "Marketplace", "partnership": "Partnership", "case_study": "Case study", "launch": "Launch", "community": "Community", "company": "Company", "trust": "Trust"}, "audienceLabels": {"users": "Users", "developers": "Developers", "investors": "Investors", "enterprise": "Enterprise"}, "proofLevelLabels": {"official": "Official source", "public-source": "Public source", "partner-approved": "Partner approved", "customer-approved": "Customer approved", "self-reported": "XMemo update"}, "proofLevelDescriptions": {"official": "Confirmed by an official platform or institution.", "public-source": "Backed by a third-party public source without implying a formal partnership.", "partner-approved": "Approved by the named partner for public display.", "customer-approved": "Approved by the named customer for public display.", "self-reported": "Self-reported XMemo product update. Use the source note to understand the public artifact or internal product boundary."}}, "hostedQuickstart": {"title": "", "body": "", "steps": []}, "memoryConsoleStory": {"title": "", "body": "", "trustHref": "", "trustCta": "", "features": []}, "faqTroubleshooting": {"title": "", "body": "", "actionLabel": "", "safetyLabel": "", "items": [], "groups": []}, "personaFlows": [], "labels": {"accountEntry": "Account entry", "agentMemoryControlPlane": "AI agent memory layer", "brand": "XMemo", "clientTargets": "client targets", "commandPalette": "Command palette", "configPath": "Config path:", "copy": "Copy", "copyAsMarkdown": "Copy as Markdown ▾", "copyConfig": "Copy config", "copyUnavailable": "Copy unavailable", "copied": "Copied", "credentialHandoff": "credential handoff", "docs": "Docs", "docsDesc": "Hosted XMemo quickstart for people connecting MCP clients.", "docsHeader": "Docs header", "docsLead": "Start with what XMemo does for you, then connect the hosted memory layer to the client you use.", "docsVersion": "Hosted · Beta pilot", "inThisArticle": "In this article", "invitedOnly": "invited only", "login": "Log in / Sign up", "mcp": "MCP", "mcpDesc": "Configure ChatGPT, VS Code / GitHub Copilot, Copilot CLI, Codex, Gemini CLI, Cursor, and custom MCP clients.", "mcpNpmQuickSetup": "⚡ Quick setup: run npx @xmemo/client setup --url https://xmemo.dev<\/code> to auto-generate per-client instance IDs and apply reviewed config for supported clients.", "mcpSetupConsole": "MCP setup console", "mcpSetupDescription": "Pick the MCP client to see the public-safe snippet for its settings surface. OAuth snippets omit bearer tokens; direct snippets reference XMEMO_KEY<\/code> only when needed plus generated, non-secret agent identity headers instead of literal credentials.", "canonicalCapabilityMatrix": "Open the canonical client capability matrix", "mcpDocsRedirect": "MCP setup details, client snippets, and tool explanations are organized in Docs.", "navTrust": "Trust", "noMatches": "No matches.", "openAccountEntry": "Open account entry ↗", "openDocsChapter": "Open Docs chapter", "openProductNavigation": "Open public navigation", "pasteSnippet": "Paste the selected snippet into the listed config path or dashboard, keep the hosted URL as shown, and set XMEMO_AGENT_INSTANCE_ID once per local client when the snippet references it. Then complete OAuth for OAuth clients or restart the direct MCP client.", "personaConnection": "Connection", "personaFlowsBody": "Choose the path that matches how you want to use XMemo. Each flow keeps credentials out of public docs and ends with a concrete memory action.", "personaFlowsTitle": "Persona onboarding flows", "personaGoal": "Goal", "personaSuccessAction": "Success action", "personaTroubleshooting": "Troubleshooting", "perClientMcpConfigPreview": "Per-client MCP config preview", "pilotAccess": "Pilot access", "product": "Home", "productDesc": "Back to the XMemo overview.", "productHome": "Home", "publicNavigation": "XMemo public navigation", "publicResources": "XMemo public resources", "loadingProductContext": "Loading product context", "productContextUnavailable": "Product context unavailable", "cachedPublicResourceRoute": "Showing the cached public resource route while the context API recovers.", "publicSafeNote": "Public-safe page: only hosted public URLs are shown; no console links, credential values, or internal control-plane endpoints are exposed here.", "quickSetup": "Quick setup", "quickstart": "Quickstart for XMemo", "search": "Search", "searchDocs": "Search docs", "searchPlaceholder": "Search docs, MCP, skills, install…", "stories": "Stories", "storiesDesc": "Milestones, ecosystem launches, and source-backed XMemo updates.", "skills": "Skills", "skillsDesc": "Connect XMemo as an agent Skill in addition to MCP.", "skillsDocsRedirect": "Skill connector guidance and agent command patterns are organized in Docs.", "start": "Get started", "startPilot": "Start pilot", "startPilotDesc": "Jump to the public pilot path.", "supportedMcpClients": "Supported MCP clients", "typeToFilter": "Type to filter", "brandHome": "XMemo home", "language": "Language", "changelog": "Changelog", "downloadSkillPackage": "Download Skill package", "installSkillTitle": "Install the Skill", "installTool": "Download tool", "curlInstall": "curl", "wgetInstall": "wget", "macosLinuxInstall": "macOS, Linux, and WSL", "windowsInstall": "Windows PowerShell", "askAgentInstall": "Ask your agent", "inspectInstallScript": "Inspect install script", "wgetAlternative": "Alternative with wget", "posixInstallExplanation": "Use this in macOS, Linux, or WSL.", "powershellInstallExplanation": "Use this in Windows PowerShell.", "agentInstallExplanation": "Paste this into an agent you trust to follow its normal safety checks.", "copyInstallCommand": "Copy install command", "posixInstallCommand": "macOS, Linux, and WSL", "powershellInstallCommand": "Windows PowerShell", "copyPowerShellInstallCommand": "Copy PowerShell install command", "copyFailed": "Copy failed. Select the command text, then press Ctrl+C or Command+C."}, "docsMarkdown": ["# Quickstart for XMemo", "", "Hosted XMemo quickstart for connecting MCP clients, CLIs, IDEs, and autonomous agents without putting bearer tokens in public configuration.", "", "- CLI package: @xmemo/client", "- Service URL: https://xmemo.dev", "- MCP URL: https://xmemo.dev/mcp", "- Setup command: npx @xmemo/client setup --url https://xmemo.dev", "- Direct MCP token environment variable: XMEMO_KEY", "- OAuth-supported clients: VS Code / GitHub Copilot, ChatGPT / OpenAI Apps SDK, Gemini CLI, Cursor, Autonomous agent", "- Persona flows: ChatGPT user, Copilot/Codex developer, team pilot owner, autonomous agent operator", "- Autonomous agent preset: OAuth-first when supported; XMEMO_KEY fallback for headless runners; runner-generated XMEMO_AGENT_INSTANCE_ID required", "- Supported MCP clients: vscode, chatgpt, claude, copilot-cli, codex, gemini-cli, cursor, autonomous-agent, other", "- MCP chapter: per-client setup snippets and hosted streamable-http boundary", "- Skills chapter: standalone Skill runtime, device login, and direct HTTPS operations", "- SDK helpers: taskStart, taskEnd, recordDecision, recordBugFix, recordCorrection, captureEvent (recordEvent is an alias)", "- Retrieval tool references: recall (https://xmemo.dev/docs/tools/recall), search_memory (https://xmemo.dev/docs/tools/search), recall_context (https://xmemo.dev/docs/tools/recall-context)", "Choosing the right retrieval tool: recall is a lightweight, forgiving lookup; search_memory is a strict, targeted search; recall_context packages multiple bounded memories into task context.", "- Troubleshooting: docs/troubleshooting (https://xmemo.dev/docs/troubleshooting)", "- Safety guarantees: manual config review, environment-variable secrets, no token in generated URLs", "", "## Hosted XMemo quickstart", "", "Start with the managed XMemo service and one of these connection paths. Self-hosted, Docker, Supabase, and Python package setup are developer references, not the first path for hosted users.", "", "- OAuth clients: https://xmemo.dev/mcp (memory:read memory:write)", "- @xmemo/client setup: npx @xmemo/client setup --url https://xmemo.dev", "- Direct MCP with XMEMO_KEY: https://xmemo.dev/mcp", "", "## Memory Console controls", "", "XMemo is not only an MCP endpoint. The Memory Console is the user-facing place to review, correct, remove, and export memory while keeping agent attribution visible.", "", "- View and search memory", "- Edit or correct entries", "- Delete and export", "- Agent attribution", "- Privacy and credential controls", "", "## FAQ / Troubleshooting", "", "Use these answers when a user or marketplace reviewer gets blocked. Every answer includes an action step and a safety boundary."], "skillPackage": null, "skillCapability": null, "docsNavTree": null, "activeRoute": null, "activeToolReference": null}, "prefixedRoutesActive": false}