PluginBench
MCP Server
Active
Apache-2.0

Fidelis Memory MCP Server

io.github.hermes-labs-ai/fidelis-memory

Local memory for AI agents: save and retrieve your notes without forgetting the details that matter.

What is the Fidelis Memory MCP server?

Fidelis Memory is an MCP server that provides persistent local memory for AI agents like Claude and Cursor. It stores Markdown and text notes and retrieves their original wording through fast vector search, hybrid keyword-vector retrieval, or model-assisted routes—all running locally without requiring external APIs.

Fidelis solves the problem of agents losing important context across sessions. Instead of relying on summaries that omit critical details, it stores your exact notes and retrieves them when needed. The default retrieval uses local vector embeddings (via Ollama) and does not rewrite your memory, preserving the original wording and conditions that matter for decision-making.

How to install Fidelis Memory

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
Environment / auth
  • FIDELIS_PORT

    Port of the local fidelis-server process (default: 19420).

~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "fidelis-memory": {
      "command": "uvx",
      "args": [
        "fidelis-memory",
        "--from",
        "fidelis-memory",
        "fidelis",
        "mcp",
        "serve"
      ],
      "env": {
        "FIDELIS_PORT": "<YOUR_FIDELIS_PORT>"
      }
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • recall — Retrieve stored notes using fast vector search based on semantic similarity.
  • recall-hybrid — Combine keyword and vector search results without a generative LLM (zero_llm tier) for more precise retrieval.
  • store — Save text or Markdown notes to the local memory store, preserving original wording.
  • watch — Monitor a directory for files and automatically ingest them into the memory store.
  • correct — Supersede a prior memory record with a correction, maintaining historical status.
  • fetch — Retrieve a specific stored note by identifier.
  • inspect — View metadata and status of stored memories.

Use cases

  • Save project decisions and rollback conditions so agents can recall why something failed and what must be verified before retrying.
  • Store debugging notes and error contexts across sessions so agents have access to the original problem description, not just a summary.
  • Maintain a local knowledge base of procedures, configurations, and edge cases that agents can search and retrieve without external API calls.
  • Track corrections and updates to notes over time, allowing agents to see both current and historical versions of important information.
  • Enable agents to recall specific technical requirements or constraints that might be omitted from high-level summaries.

Fidelis Memory MCP server FAQ

What is Fidelis Memory?

Fidelis Memory is a local memory service for AI agents that stores and retrieves your exact notes without rewriting them. It runs on your machine and integrates with Claude, Cursor, Codex, and other MCP-compatible clients.

Is Fidelis Memory free?

Yes. Fidelis Memory is open-source under the Apache-2.0 license and runs entirely on your local machine. It requires a local Ollama service for embeddings, which is also free.

How do I install it in Cursor or Claude?

Install the Python package (`pip install fidelis-memory`), run `fidelis init` to set up the background service, then run `fidelis mcp install` (for Claude Code) or `fidelis mcp install --client codex` (for Cursor). Restart your client and the tools will be available.

What are the system requirements?

Python 3.10+, macOS or Ubuntu, and a local Ollama service. The default embedding model is nomic-embed-text. No API keys are required.

Does Fidelis rewrite my notes?

No. The default retrieval path preserves your original wording. Optional extraction and model-assisted routes exist, but the verbatim write paths (`watch`, `store`) keep the text exactly as you supplied it.

Is this a team or cloud service?

No. Fidelis 0.3.0rc1 is a single-machine pre-release. All data is stored locally at `~/.cogito/store` on your machine, not on a hosted server.

README (reference)

Source of truth, from the repository.

<!-- mcp-name: io.github.hermes-labs-ai/fidelis-memory --> <div align="center">

Fidelis Memory

<img src="assets/fidelis-memory-artwork.png" width="420" alt="Fidelis Memory mascot" />

Your agent can forget a summary. Fidelis brings back the note.

Local memory for Codex, Claude Code, and other agents: save text across sessions and retrieve the original passage when it matters.

by Hermes Labs

PyPI · Connect an agent · Technical reference

</div>

A project summary might say “billing migration rolled back.” The original note says why it failed and do not retry until the idempotency fix is verified. When your next session depends on that condition, the missing sentence matters.

Fidelis stores your Markdown and text notes and retrieves their stored wording. Its default retrieval does not ask a generative model to rewrite your memory.

Try it with one note

This path supports Python 3.10+, macOS or Ubuntu, and a local Ollama service. Start Ollama if needed (ollama serve in a separate terminal). No model API key is required. The current release is 0.3.0rc1.

ollama pull nomic-embed-text
python3 -m venv ~/.venvs/fidelis
source ~/.venvs/fidelis/bin/activate
python3 -m pip install "fidelis-memory[hybrid]==0.3.0rc1"
fidelis init

demo_dir=$(mktemp -d)
cat > "$demo_dir/atlas.md" <<'NOTE'
Atlas billing migration:
Duplicate charges appeared in staging. Rolled back.
Do not retry until the idempotency fix is verified.
NOTE

fidelis watch "$demo_dir" --once
fidelis recall-hybrid "Atlas billing migration retry condition" --tier zero_llm

Look for both the rollback and the retry condition in the retrieved text. This is retrieval of a saved note, not a generated answer. For your own files, replace "$demo_dir" with a notes directory. If retrieval fails, run fidelis health and confirm Ollama has nomic-embed-text available. Keep the virtual environment after fidelis init; the background service uses it.

What this changes on your machine: fidelis init installs a per-user background service (launchd on macOS, systemd --user on Linux) that starts fidelis-server automatically at login. fidelis watch sends the note to that server, which saves it in the local store at ~/.cogito/store by default, not in the temporary demo directory. The watcher also records ingested files in ~/.fidelis/watched.json. To remove the default service on macOS, run fidelis init --uninstall; it stops the service and deletes the launchd plist only. On Linux, find the installed unit with systemctl --user list-unit-files '*fidelis-server*', then run systemctl --user disable --now <unit>, delete that unit file from ~/.config/systemd/user/, and run systemctl --user daemon-reload. Neither path deletes your data. The store is shared by everything you save with Fidelis, so deleting ~/.cogito/store erases all of it, not just the demo note; do that only on a throwaway install. Also delete the temporary demo directory ($demo_dir), and ~/.fidelis/watched.json if you want its record of the file path gone.

Connect an agent

Once the note is retrievable locally, install the MCP connection for your client:

fidelis mcp install --client codex
# Or, for Claude Code:
fidelis mcp install

Restart the client and try: “Use Fidelis to find my Atlas billing migration note. What must happen before we retry? Quote the relevant text.”

The client gets tools to recall, store, correct, fetch, and inspect memory. It decides when to call them; installing the connection does not make every turn recall automatically. Other clients and uninstall commands include Cursor, Gemini CLI, Copilot CLI, and OpenClaw. A separate Pi extension provides prompt-time recall.

What Fidelis preserves

  • Original wording. watch, store, and other verbatim write paths keep the text you supplied, so retrieval can return the condition that a summary might omit.
  • Local retrieval. Fast vector search is the default MCP recall path. Explicit hybrid search combines keyword and vector results without a generative LLM at its zero_llm tier; local embeddings still require Ollama.
  • Corrections over time. A correction can supersede a prior record without erasing it. Recall can show validity and historical status instead of silently replacing the old note.

Fidelis also has optional extraction and model-assisted routes. Those can transform or filter inputs; use the verbatim path when the original wording is the point. Retrieved text can still be wrong or out of date, and your agent's model provider may see it when answering.

Current scope

Fidelis 0.3.0rc1 is a single-machine pre-release, not a hosted team memory service. The LongMemEval-S retrieval run measured whether the redesigned default path retrieved material across 470 questions; it did not measure answer accuracy. See the results, user-fit guide, and security policy for deeper detail.

Full reference · Upgrade notes · Contribute · Changelog

Apache-2.0. Hermes Labs builds agentic infrastructure for autonomous systems.

Related MCP servers

HEHermes Prime logo

Read-only stdio MCP server for a versioned convention card in Claude Code.

2
Shell
MIT
View repository →

MCP server for SQL Server Express LocalDB on Windows: manage instances, run T-SQL, discover DBs

1
Rust
MIT
View repository →

Task manager for AI agents. Zero setup, nested trees, batch ops, local-first storage.

View repository →
HEHermoso logo

Hermoso

Active

Marketing on autopilot from your agent. 864 tools, usable alone. Publishing and ads cost no credits.

0
JavaScript
MIT
View repository →

Search Argentine Supreme Court sumarios: legal topics, case analysis, metadata

Search Buenos Aires Province court rulings: summaries, full text, legal topics