Skip to content
Agentmart
Submit a tool

Fidelis Memory

Recall locally stored agent memories through an MCP-compatible host.

Connector by hermes-labs-ai · GitHub account verified by the MCP Registry

IndexedApache-2.0Local (stdio)★ 24
Write a review

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/hermes-labs-ai-fidelis-memory

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Recall locally stored agent memories through an MCP-compatible host.

Use cases: agent-memory, knowledge-base

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools
4 read · 1 write · 0 delete · 1 not yet classified
Secrets
None asked for
Install scripts
None found
Runs as
Python package

Tools by what they do 6

Read from the source code; the running server may list different tools.

Read 4 · Look things up without changing anything.

  • fidelis_get
  • fidelis_health
  • fidelis_recall
  • fidelis_recent

Write 1 · Create, change, send or run things.

  • fidelis_store

Not yet classified 1 · We could not tell from the name and description.

  • fidelis_correct

Secrets and settings it asks for

Other settings (environment variables or headers).

  • FIDELIS_PORT

Network hosts it talks to 15

  • agent-plugins.org · found in code
  • api.anthropic.com · found in code
  • api.openai.com · found in code
  • apple.com · found in code
  • dashscope-intl.aliyuncs.com · found in code
  • github.com · found in code
  • glama.ai · found in code
  • hermes-labs.ai · found in code
  • huggingface.co · found in code
  • ollama.com · found in code
  • orcid.org · found in code
  • source.unsplash.com · found in code
  • speyer.de · found in code
  • universalstudioshollywood.com · found in code
  • w3id.org · found in code

Files it touches

  • /tmp/not-mcp_server.py-backup
  • Your home directory (~)
  • ~/.cogito/escalation.log
  • ~/.cogito/store
  • ~/.copilot
  • ~/.copilot/mcp-config.json
  • ~/.gemini/settings.json
  • ~/.openclaw/openclaw.json
  • Any path the agent passes to its tools

Tools (6)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • fidelis_correct

    Correct a memory that is wrong or outdated, in one call: stores `text` as a new record superseding `id`. The old record is kept, marked superseded, and stays visible in fidelis_get's history -- nothing is deleted. `id` must come from a prior fidelis_recall or fidelis_get result; an unknown id is rejected. If `id` is already superseded, this reports the current replacement instead of writing, unless force=true. Do not use this for a brand-new, unrelated fact -- use fidelis_store instead.

  • fidelis_get

    Fetch exactly one memory by id, verbatim, with its correction chain in both directions: what it replaced (oldest last) and what replaced it (newest last). Use this to resolve a "SUPERSEDED by <id>" marker from fidelis_recall, or to confirm an id before passing it to fidelis_correct. An unknown id is an error. Do not use this to find a memory by topic -- use fidelis_recall instead.

  • fidelis_health

    Check whether the memory backend is up, and see store-wide stats (current/superseded/expired counts, oldest/newest record, queue and dead-letter depth) in one call. Call this after any other fidelis_* tool returns isError, before retrying. Do not call this to search memory -- use fidelis_recall for that.

  • fidelis_recall

    Search stored memories for `query`. Call before answering questions about prior work, decisions, or past conversations. Default mode="fast" (vector search, ~0.1s); use mode="thorough" only when a fast result looks wrong or stakes are high (~3s, hybrid search). Current facts list first, superseded/expired after, marked. Every hit shows its id, status, and recorded date. Fix a stale hit with fidelis_correct; see its history with fidelis_get. Do not use this to store a new fact -- use fidelis_store instead.

  • fidelis_recent

    List memories by recency, no search query needed -- good for "what changed recently" or catching up after time away. `since` (ISO-8601) is a lower bound: only records at or after it. kind="corrections" shows only memories that corrected an older one, with the id(s) they replaced. Do not use this to find a memory about a topic -- use fidelis_recall instead.

  • fidelis_store

    Remember or store one new, self-contained memory verbatim -- exact text, not a summary. Optional valid_from/valid_to/event_at/source, and supersedes (ids this replaces; unknown or another user's ids are rejected). The reply echoes recorded_at and any declared window so you can confirm without a recall. Do not use this to correct something already in memory -- use fidelis_correct with the old id instead of storing a near-duplicate.

Signals

Stars
24
Forks
3
Open issues
4
Contributors
4
Weekly downloads
66
Last commit
Oct 19, 2026 (9 days ago)
Last release
Sep 20, 2026 (1 month ago)
Last package publish
Sep 20, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name
io.github.hermes-labs-ai/fidelis-memory
Version
0.2.0
Auth
No auth
Packages
pypi fidelis-memory
License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from hermes-labs-ai

Cite this page

Fidelis Memory MCP server. Agent Mart. https://agentmart.dev/connectors/hermes-labs-ai-fidelis-memory (last updated 2026-10-24)