--- title: Phasoric | AI Agent Memory for ChatGPT, Claude & Codex description: Give Codex, Claude, ChatGPT, Cursor and compatible AI agents durable access to authorized project knowledge, decisions and history through Phasoric. canonical_url: https://phasoric.com/ai-agent-memory markdown_url: https://phasoric.com/ai-agent-memory.md --- In partnership with every AI you use # Stop teaching your AI the same project twice. Keep your notes, research, and decisions in Phasoric. Connect ChatGPT, Claude, Codex, Cursor or another compatible AI client, then ask it to retrieve the knowledge you authorize. Save reviewed conclusions so your next session can build on them. Use it through MCP Connect your AI tools to authorized knowledge. Use hosted intelligence Ask Phasoric directly. No MCP setup required. [Start with Premium — $14.99/mo](/signup?target=ai-agent-memory&plan=premium&interval=month) [Try hosted intelligence](/hosted-ai-memory?target=ai-agent-memory) Premium includes 2,000 MCP operations and 25,000 hosted-intelligence credits per month. Portable Markdown · Scoped access · Review before writes The knowledge you choose to carry forward Plans · research · decisions · lessons learned Phasoric Connected knowledge, with sources and permissions ChatGPT / Claude / Codex Cursor · GitHub Copilot · compatible MCP clients Muse · Grok Bot · ChatGPT Dots, where compatible MCP connections are available “What did we decide?” “What changed?” “Write this decision back.” Carry what you learn between AI tools ## Make the AI you already pay for more useful. Keep using Codex, Claude, ChatGPT or Cursor for the work you chose them for. Phasoric supplies the project context: the architecture decision, customer constraint, rejected approach or research you would otherwise explain again. Save what matters from one session so the next can use it. Keep that learning in a reviewed project record, then ask your agent to retrieve it when you need it. [Explore coding agent memory](/ai-coding-agent-memory?target=ai-agent-memory)[See our real Codex workflow](#workflow) From our own work · September 2026 ## A plan in the Vault became a working change. We asked Codex to read a Phasoric Vault brief about persistent MCP connections. It retrieved the plan through Phasoric, investigated the OAuth flow, and implemented the fix in the repository. The reasoning traveled with the task. 1. 01 ### Read the brief An existing Markdown plan supplied the problem, constraints, and acceptance criteria. 2. 02 ### Investigate and implement Codex used the brief to guide the code changes and tests in the project. 3. 03 ### Review and merge The fix was merged into develop after approval, with release validation still explicit. Explore the workflow What did we decide?What changed?Write this decision back. Read the context. Do the work. Save the reasoning for next time. Recover the rationale Keep access tokens short-lived. Use rotating refresh tokens so an authorized agent can renew access silently. Keep vault-provider credentials behind Phasoric. Source: Persistent MCP OAuth brief · requirements and constraints The agent reads the original plan and its reasoning before proposing a change. Inspect the example ## A new constraint. A decision worth revisiting. Fictional project, prepared sources and review. These controls reveal the example; they make no AI requests, save no notes and grant no permissions. 1\. Original decision2\. Changed evidence3\. Review outcome ### Original decision Atlas/Decisions/014-import-queue.md Revision: synthetic-v1 · Recorded: 2026-09-01 Keep imports asynchronous so a browser closing does not interrupt the job. Retain the original upload for seven days. Reuse a stable job key when retrying. The seven-day retention period is an assumption to review if storage policy changes. ### What the next session needs - The original reason for the queue. - The new limit on retention. - The unresolved recovery question. - The fact that a proposal is not approval. A real client must retrieve these records from an authorized mapping and inspect the current revisions. Inspect the context-selection case Question: Can I replace the queue with a longer browser timeout? Expected observation: A retention change does not remove the need to survive a browser closing. Cite both sources, preserve the queue requirement and flag the unresolved recovery question. Excluded: an unrelated project launch note. These expected observations are authored examples. No retrieval algorithm, authorization boundary or model was executed to produce them. [Download the complete synthetic example (JSON)](/assets/decision-review-example-Dn-3Ig09.json) ## Inspect the contract behind the workflow. The public MCP reference describes authorized mappings, decision evidence and write previews. Documentation explains the contract; the prepared example above is not a connected test or a measured result. [Read the public MCP contract and client configuration](https://developers.phasoric.com/mcp?target=ai-agent-memory)[Review methods and limitations](/benchmarks?target=ai-agent-memory) Documented client setup ## Start in the tool you already use. [ChatGPTBring project context into your conversations.Setup guide](/developers/mcp-chatgpt?target=ai-agent-memory) [ClaudeUse a remote connector, or connect Claude Code.Setup guide](/developers/mcp-claude?target=ai-agent-memory) [CodexRead the brief behind the implementation.Setup guide](/developers/mcp-codex?target=ai-agent-memory) [CursorKeep engineering rationale beside your code.Setup guide](/developers/mcp-cursor?target=ai-agent-memory) [VS Code / GitHub CopilotGive agent mode access to your project knowledge.Setup guide](/developers/mcp-vscode?target=ai-agent-memory) An authenticated Vault read was verified in Codex. The other guides describe documented setup paths; client plans, versions and workspace policies determine availability. They do not establish Phasoric interoperability for every client version. [Prepare your vault and verify a first read](/developers/mcp-client-setup?target=ai-agent-memory) ### More remote MCP setup paths OpenClaw, Hermes Agent and Grok document remote MCP connection paths. These guides map those instructions to Phasoric; authenticated interoperability has not been verified for these clients. [OpenClaw setup](/developers/mcp-openclaw?target=ai-agent-memory)[Hermes Agent setup](/developers/mcp-hermes?target=ai-agent-memory)[Grok setup](/developers/mcp-grok?target=ai-agent-memory) Durable context. Deliberate access. ## Keep control of the record. Phasoric exposes the knowledge you authorize. It does not silently ingest every conversation or automatically remember everything your agents do. [Inspect the MCP boundaries](/mcp-knowledge-base?target=ai-agent-memory) - ### Choose the source Hosted MCP reaches explicitly mapped Phasoric Hosted, Dropbox, Google Drive or GitHub vaults. Local-only vaults and device-encrypted remote bodies remain unavailable. - ### Keep the provenance Ask for the note path, evidence and limits of the answer. Your Markdown remains portable across tools. - ### Approve the write Review a proposed note change before application. Write permission, current-content checks and duplicate-request safeguards still apply. ## Before you begin Is this the same as an AI chat history? Phasoric gives your agent access to an explicitly connected knowledge source: Markdown notes, plans, and decision records that persist beyond a conversation. The agent retrieves relevant material when you ask; it does not automatically remember every chat or ingest every file. Can I use my existing Markdown vault? Yes, when it is explicitly mapped through Phasoric Hosted, Dropbox, Google Drive, or GitHub and your account allows access. A local-only browser vault is not reachable by hosted MCP. Device-encrypted remote content remains unavailable to the hosted service. Can an agent change my notes? Yes. Your agent can write to an authorized connected Vault after you review and approve the change. Safeguards protect against stale content and duplicate requests. You can use read-only access instead. Which plans and clients do I need? Hosted MCP starts with Phasoric Premium; higher plans have higher usage limits. Your agent also needs remote MCP support and any permissions required by its vendor or workspace. The setup guides explain the client-specific steps; a guide is not a claim that every version or plan has been tested. Start with one project ## Give the next session something to build on. Keep the brief, recover the evidence and save the reviewed decision. Choose how you want to use that context. ### Use it through MCP Connect the AI tools you already use to your authorized Phasoric knowledge. Hosted MCP starts with Premium. $14.99/month includes 2,000 MCP operations and 25,000 hosted-intelligence credits each month. Higher plans increase limits and features. [Connect my AI tools](/signup?target=ai-agent-memory&plan=premium&interval=month) ### Use hosted intelligence No MCP setup required. Ask Phasoric directly using hosted intelligence with bounded relevant source context. Eligible Free hosted accounts can buy 50,000 credits for $5.99. Usage is metered; a credit pack does not enable MCP. [Try hosted intelligence](/hosted-ai-memory?target=ai-agent-memory) [Coding agent memory](/ai-coding-agent-memory?target=ai-agent-memory) [Hosted AI memory](/hosted-ai-memory?target=ai-agent-memory) [MCP memory & knowledge base](/mcp-knowledge-base?target=ai-agent-memory) [AI memory for your Obsidian vault](/obsidian-ai-memory?target=ai-agent-memory) [Context between AI tools](/ai-context-handoff?target=ai-agent-memory) [AI projects & durable context](/compare/ai-project-memory?target=ai-agent-memory) [Compare all plans](/pricing?target=ai-agent-memory)