Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

io.github.writerslogic/cogmem

PYPI · COGMEM · SCANNED SEP 20

A self-improving, verifiable memory layer for AI coding agents.

Available components

0 this week 77 Trust /100
Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →

Supply Chain Security100
  • No malware found by supply-chain analysis.Pass
  • No known CVEs affecting this package version or its production dependencies.Pass
  • Runs setuptools.build_meta at install time, a recognised native-build step with no shell scripting around it. View diagnostics → Pass
  • 0 of 5 dependencies flagged as unhealthy. View diagnostics → Pass
Provenance & Transparency10
Schema Quality & AI Usability87
  • 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
  • AI-judged instruction clarity (excellent).Pass
  • Tool/resource definitions use about 746 tokens (~82/item across 9 items; 8 tools + 1 resources), lean.Pass
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management83
  • Stability observed for 25 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • We read all 8 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 9 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Install

How do I install the io.github.writerslogic/cogmem MCP server?

io.github.writerslogic/cogmem runs locally as a PyPI package, launched with uvx cogmem. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

pypi · cogmem

# add to Claude Code
claude mcp add writerslogic-cogmem -- uvx cogmem
// .cursor/mcp.json
{
  "mcpServers": {
    "writerslogic-cogmem": {
      "command": "uvx",
      "args": [
        "cogmem"
      ]
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "writerslogic-cogmem": {
      "command": "uvx",
      "args": [
        "cogmem"
      ]
    }
  }
}
# add to Codex CLI
codex mcp add writerslogic-cogmem -- uvx cogmem
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "writerslogic-cogmem": {
      "type": "local",
      "command": [
        "uvx",
        "cogmem"
      ],
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add writerslogic-cogmem --command uvx --arg cogmem
# ~/.hermes/config.yaml
mcp_servers:
  writerslogic-cogmem:
    command: "uvx"
    args: ["cogmem"]
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "writerslogic-cogmem": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "cogmem"
      ]
    }
  }
}
# add to Vellum
assistant mcp add writerslogic-cogmem -t stdio -c uvx -a cogmem
// mcp.json
{
  "mcpServers": {
    "writerslogic-cogmem": {
      "command": "uvx",
      "args": [
        "cogmem"
      ]
    }
  }
}
Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 19 Sept 26 −3
    • Stability: pass → 0.80 functional
  • 18 Sept 26 +1
    • Stability: 0.97 → pass security
  • 16 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.

  • 14 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

  • 13 Sept 26 −3
    • Stability: pass → 0.83 functional
  • 12 Sept 26 +1
    • Stability: 0.97 → pass security
  • 10 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.

  • 8 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 20 Sept 2026 · Analysed pypi/cogmem@2.7.1

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem pypi

Background: How many MCP packages publish verified provenance →

Install scripts 1 script
Hook Tier Command
build_backend allowlisted setuptools.build_meta

Background: Why install scripts are a supply-chain risk →

Dependencies 5 packages
Packages resolved 5
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 8 exposed · ~732 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
note ~112

Record a decision, finding, or correction into memory mid-task so it can be recalled in future sessions. Returns {ok, noted}. The text is captured as a candidate and deduped against existing knowledge by the background pipeline. Use when the user states a durable preference or you learn something worth keeping; not for transient chatter. To retrieve memories use `recall`.

NameTypeReqDescription
textstringyesThe lesson to remember, as one self-contained sentence, e.g. 'The user prefers Conventional Commits with no body.'
NameTypeReqDescription
notedstring
okbooleanyes

No examples provided.

progress ~54

Summarize momentum, stalls, and dependencies across the user's projects as a narrative. Returns {narrative}. Use to orient at session start or when the user asks 'where are we'. Read-only synthesis of project-state memory.

Input schema present but exposes no named parameters.

NameTypeReqDescription
narrativestringyes

No examples provided.

recall ~172

Surface the most relevant past lessons, decisions, and rules for a task, ranked by semantic similarity. Returns {count, memories:[{id, scope, score, text}]}, where score is rerank confidence (higher = more relevant). Use at the start of a task or whenever unsure how the user wants something done, instead of guessing. Read-only — to save a new memory use `note`.

NameTypeReqDescription
kintegerMaximum number of memories to return (1-20). Defaults to 5.
querystringyesNatural-language description of the task or question to find lessons for, e.g. 'how does the user want commit messages formatted'.
scopestringOptional domain filter, e.g. 'rust', 'python', 'universal'. Omit to search every scope.
NameTypeReqDescription
countintegeryes
memoriesarrayyes

No examples provided.

receipt ~103

Produce an RFC 6962-style cryptographic proof that a specific memory is committed in the signed transparency log. Returns the inclusion receipt (leaf index, audit path, tree size, signed root). Use to prove to a third party that a memory existed and was logged. Requires the memory's id — get ids from `recall`.

NameTypeReqDescription
memory_idstringyesThe id of the memory to prove inclusion for, as returned in a recall result's `id` field.

Structured output declared, but exposes no named fields.

No examples provided.

review_pending ~61

List always-load (Layer-A) rules awaiting human approval before they enter the always-on context. Returns {pending}. Use to see what the system wants to promote. Approval itself is a human action via the `cogmem review` CLI, not this tool.

Input schema present but exposes no named parameters.

NameTypeReqDescription
pendingstringyes

No examples provided.

status ~90

Report the health of the verifiable-memory system. Returns {agentDid, logEntries, logIntegrity, merkleRoot} — the agent's did:key identity, transparency-log size, its integrity ('ok' or a reason), and the current Merkle root. Use for a fast health/identity check. For a full per-memory credential audit use `verify`; for the signed log commitment use `tree_head`.

Input schema present but exposes no named parameters.

NameTypeReqDescription
agentDidstringyes
logEntriesinteger
logIntegritystringyes
merkleRootstringyes

No examples provided.

tree_head ~66

Return the current signed Merkle tree head — the log's tamper-evident commitment to every memory so far. Returns {rootHash, treeSize, signature, ...}. Use as the anchor a verifier checks inclusion receipts against, or to detect log forks. Pair with `receipt`.

Input schema present but exposes no named parameters.

NameTypeReqDescription
rootHashstring
treeSizeinteger

No examples provided.

verify ~74

Cryptographically verify every stored memory's W3C Verifiable Credential and the integrity of the hash-chained transparency log. Returns a summary of memories checked, how many are valid, and any failure reasons. Use to detect tampered or poisoned memories before trusting them. This is the deep audit; `status` is the lightweight check.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

Common questions

What is the io.github.writerslogic/cogmem MCP server?

io.github.writerslogic/cogmem is an MCP server listed in the public MCP registry as io.github.writerslogic/cogmem. A self-improving, verifiable memory layer for AI coding agents. This page covers its PyPI package (cogmem).

Is the io.github.writerslogic/cogmem MCP server safe to use?

io.github.writerslogic/cogmem scores 77 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the io.github.writerslogic/cogmem MCP server expose?

io.github.writerslogic/cogmem exposes 8 tools: recall, note, status, verify, receipt, and 3 more. Their descriptions and schemas cost roughly 732 tokens of context every time the server is loaded.

Is the io.github.writerslogic/cogmem MCP server still maintained?

io.github.writerslogic/cogmem is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.