RE-call MCP Memory Server MCP server

by GiulioDER·io.github.GiulioDER/re-call·v0.14.0·6 stars

No reviews yet

Agent memory on your own Postgres that flags superseded facts and abstains instead of guessing.

B72/100grade B

Write a review Install

What users say

No reviews yet

Be the first

Safety scan

B72/100

full report

Adoption

Growing

6 stars77 downloads/wk

Reviews

Write one

Nobody has reviewed RE-call MCP Memory Server yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

RE-call MCP Memory Server tools (23, 5 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

__init__
recall_apply_fact
Apply one structured atomic fact through the deterministic provenance controller.
recall_calibration_publishwrite action
Publish one certified calibration artifact for the caller's tenant.
recall_calibration_runwrite action
Create a draft calibration artifact for the active generation.
recall_calibration_status
Return the latest calibration artifact bound to the caller's generation.
recall_current_facts
Return the tenant scoped current projection of the append only fact ledger.
recall_current_state
Return a bounded, deterministic, generation bound authored state projection.
recall_evidence
Get memory as CITABLE EVIDENCE plus the exact prompt to answer it with.
recall_forget
Permanently delete indexed memory for the given source(s). IRREVERSIBLE.
recall_index
Index a markdown file or folder into the agent's memory so it can be recalled later.
recall_ingestwrite action
Upload bounded source files and index them in the caller's tenant.
recall_inventory
List every source in memory with its raw content digest for client-side sync.
recall_job_status
Return the current state of one bounded indexing job.
recall_query_construction_challenge
Ask the original model for a bounded frame, then continue retrieval.
recall_reasoning_auditwrite action
Run the bounded integration audit without disclosing corpus or query text in errors.
recall_reasoning_projection
Inspect the immutable reasoning projection for this tenant and generation.
recall_reasoning_proposals
List side effect free inference proposals for human review.
recall_reasoning_querywrite action
Run explicit opt-in reasoning over trusted retrieval and a derived graph.
recall_related
Return independently trusted structural neighbors of one evidence chunk.
recall_rewrite_plan
Show which key a proposal would declare, in which file. Writes nothing.
recall_search
Search the agent's OWN memory before acting, and get actionable guidance.
recall_stats
Report how much memory exists and whether it is stale (freshness check).
recall_tenants
Return the tenant scopes visible to this caller.

Public scan report

scanner v0.1.10 · 2026-10-15 · same rubric, same numbers if you re-run it

1 medium

Code scan222 source files scanned20/25
–Live reliabilityno gateway calls yet and no remote to proben/a
–Tool poisoningtools not inspected (local package is not executed); not countedn/a
Auth qualitystatic API keys via environment variables6/15
Maintenancelast push 0 days ago15/15
Maintainer identityregistry namespace matches repository owner6/10

Findings (1)

mediumeval / new Function usedexec.eval

x/recall/sparse.py: …m_pretrained(model_name, **loader_args).eval() # `from_pretrained` loads to …

Overall 72/100. Components that don't apply are left out of the denominator. Any critical finding is an F.Rubric Appeal a finding JSON

How RE-call MCP Memory Server compares

Against every graded Databases MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score: 72/100; higher than 30% of 802 Databases servers
Tools: 23; the typical Databases server has 8
Tools that change things: 22%; typical: 10%. Fewer write tools means less an agent can break
GitHub stars: 6; more than 75% of 602 Databases servers
Downloads a week: 77; more than 50% of 328 Databases servers with a package

What its tools let an agent do: 18 read · 5 change

Release history: 2 versions in the registry since 2026-09-09, latest 2026-09-25, typically one every 16 days.

Install RE-call MCP Memory Server in Claude Code, Cursor or VS Code

claude mcp add re-call -- uvx recall-rag

Add to Cursor

What the publisher says

From the RE-call MCP Memory Server repository's README, as published. We do not edit it. Read it on GitHub

Memory that abstains instead of guessing. RE-call is agent memory on your own PostgreSQL with pgvector. Each result carries a verdict, confidence, and provenance, while unsupported questions are refused instead of answered by the nearest match.

What it is  ·  For agents  ·  How it works  ·  Try the demo  ·  Install and integrate

What is RE-call

RE-call is agent memory on your own PostgreSQL database. It indexes source documents with pgvector and keeps validity, lineage, confidence, and provenance attached to every result.

Vector search returns nearby text. RE-call also checks whether that text is current, supported, and trustworthy enough for the query. A superseded claim is marked superseded; a result that does not clear the calibrated trust gate becomes ABSTAIN with a reason. Declared supersession makes the current memory win over stale but similar memory.

The default path uses local embeddings plus hybrid dense and Postgres full text retrieval. It needs no memory-layer LLM call. Additional retrieval, reasoning, and structured-fact modules are optional.

For the agent reading this, RE-call for agents explains the experience loop and how to use persistent memory as a trustworthy past rather than an unverified scratchpad.

How it works

RE-call builds an immutable, calibrated generation once and serves queries against that pinned generation. The solid path is the default. Dashed arrows mark the optional reasoning graph path or generation bindings. Nothing bypasses the generation or trust boundary.

flowchart TD

subgraph row_top[" "]
  direction LR
  subgraph group_clients["Entry points"]
    node_cli["CLI<br/>command interface<br/>[cli.py]"]
    node_package_api["Python package API<br/>library API<br/>[__init__.py]"]
    node_agent_sdk["Agent memory SDK<br/>in-process SDK<br/>[memory.py]"]
    node_mcp_server["MCP server<br/>tool server<br/>[server.py]"]
  end
end

subgraph row_middle[" "]
  direction LR
  subgraph group_serving["Retrieval and trust"]
    node_retriever["Hybrid retriever<br/>query service<br/>[retriever.py]"]
    node_retrieval_legs["Dense, FTS, sparse legs<br/>retrieval components<br/>[embeddings.py]"]
    node_reranker["Reranker<br/>candidate ranking<br/>[rerank.py]"]
    node_trust_gate{{"Trust gate<br/>policy enforcement<br/>[trust.py]"}}
    node_calibration["Calibration<br/>generation readiness<br/>[calibration.py]"]
  end

  subgraph group_build["Generation build"]
    node_setup_wizard["Setup wizard<br/>provisioning workflow<br/>[setup.py]"]
    node_generation_builder["Generation builder<br/>immutable indexing pipeline"]
    node_document_ingest["Document ingestion<br/>document processing<br/>[document.py]"]
    node_indexer["Index writer<br/>indexi

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

RE-call MCP Memory Server: common questions

Is RE-call MCP Memory Server MCP server safe?: Mostly: it is graded B (72/100). Read the RE-call MCP Memory Server safety report
How do I install RE-call MCP Memory Server?: It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does RE-call MCP Memory Server need an API key?: Yes. The registry entry asks for RECALL_SERVING_DSN.
Is RE-call MCP Memory Server maintained?: The last commit was 20 days ago (2026-10-14). The latest release is v0.14.0.
What can I use instead of RE-call MCP Memory Server?: Servers from other publishers that do the same job: Run402 MCP server, Memory Vault MCP server and CogniCore MCP server.
How does RE-call MCP Memory Server compare with other Databases MCP servers?: safety score 72/100 (higher than 30% of 802 Databases servers); tools 23 (the typical Databases server has 8); tools that change things 22% (typical: 10%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to RE-call MCP Memory Server

Same job from other publishers: the closest match first, then the best rated.

Run402
x402 pay-per-call infra for agents: $0.03 image generation, Postgres, auth, storage, functions.
not reviewed
Established
B
Memory Vault
Local-first AI memory layer with hybrid search. Postgres + pgvector. Self-hosted, MIT.
not reviewed
Growing
B
CogniCore
Memory, reflection and safety for AI agents. Hosted or local, no vector database.
5.0(1)
Growing
A
Appwrite
Build and deploy backends: auth, Postgres/MySQL/vector databases, storage, functions, hosting, CDN
not reviewed
Established
A
Ekwo OS — open source accounting: ledger, invoices and VAT on your own Supabase or Postgres
Double-entry books an agent keeps as the user: invoices, VAT returns, 174 country packs.
not reviewed
Established
A

Run it through the gateway

Local package. It runs on your machine, not through the gateway.

Rating: —
Safety grade: B · 72/100
Stars: 6
Downloads/wk: 77
Price: Free
Licence: AGPL-3.0
Transport: Local · pypi
Version: 0.14.0
Category: Databases
Last release: 2026-10-14

Credentials it needs

RECALL_SERVING_DSNsecretPostgreSQL serving DSN for the RE-call memory store.
RECALL_TENANTTenant id served by this local stdio server.
RECALL_TRUST_MODESet to development only for local uncalibrated corpora.
RECALL_EMBEDDERThe embedder the corpus was BUILT with, for example voyage:voyage-4 or BAAI/bge-large-en-v1.5. It must match: several models emit 1024 dimensions, so a mismatch does not error, it returns a confidently ranked list that means nothing.
RECALL_TABLEThe chunk table inside that database. `recall quickstart` indexes into quickstart_chunks so its sample corpus can never be retrieved beside real memory; `recall setup` uses chunks. The wrong value returns no hits and no error.
RECALL_MCP_TOOLSOptional. Serve a subset of the tools: a preset (all, search, read) or explicit names. Every tool definition is re-sent every turn, so serving ones nobody calls costs input tokens on every turn. Unset serves them all.

Own this server?

Claim it to keep the listing current and reply to reviews. Claiming never changes the grade or hides a review.

Claim listing

Badge for your README

Shows the safety grade and links back to this page. It updates itself after every scan and review.

[![mcp.market](https://mcp.market/badge/giulioder-re-call.svg)](https://mcp.market/server/giulioder-re-call)

How the badge works