# ai.trags/poppy (pypi · poppy-memory)

Developer memory CLI and MCP server. Remember decisions, preferences, and lessons locally.

- Trust score: 81/100 (high trust)
- Change this week: +5
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- pypi · `poppy-memory`: 81/100 (this document), [markdown](https://verifymcp.io/servers/ai-trags-poppy/poppy-memory.md), [page](https://verifymcp.io/servers/ai-trags-poppy/poppy-memory)

## Channel facts

- Registry: `pypi`
- Package: `poppy-memory`
- Version: `0.3.6`
- Transport: `stdio`

## Trust breakdown

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. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Supply Chain Security**: 99/100
  - No malware found by supply-chain analysis.
  - Known CVEs were checked across the 88 of 90 dependencies we could resolve, so this covers what we could see, not the whole tree.
  - Runs hatchling.build at install time, a recognised build step with no custom scripting around it.
  - Dependency health was assessed across the 88 of 90 dependencies we could resolve, so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to trags-garden/poppy).
  - Clear OSI-approved license (Apache-2.0).
  - Actively maintained (last published 2 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 67/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 914 tokens (~114/item across 8 items; 8 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 33/100
  - Stability observed for 10 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 71/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "forget" implies "delete" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 8 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the ai.trags/poppy MCP server?

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

### Claude

```bash
claude mcp add ai-trags-poppy -- uvx poppy-memory
```

### Cursor

```json
{
  "mcpServers": {
    "ai-trags-poppy": {
      "command": "uvx",
      "args": [
        "poppy-memory"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "ai-trags-poppy": {
      "command": "uvx",
      "args": [
        "poppy-memory"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add ai-trags-poppy -- uvx poppy-memory
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-trags-poppy": {
      "type": "local",
      "command": [
        "uvx",
        "poppy-memory"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-trags-poppy --command uvx --arg poppy-memory
```

### Hermes

```yaml
mcp_servers:
  ai-trags-poppy:
    command: "uvx"
    args: ["poppy-memory"]
```

### Netclaw

```json
{
  "McpServers": {
    "ai-trags-poppy": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "poppy-memory"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add ai-trags-poppy -t stdio -c uvx -a poppy-memory
```

### Other

```json
{
  "mcpServers": {
    "ai-trags-poppy": {
      "command": "uvx",
      "args": [
        "poppy-memory"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-05 (score 81, +20)

- [security improvement] Malware scan: unverified → pass
- [functional improvement] Stability: unverified → 0.33

### 2026-11-03 (score 61, −15)

- [security regression] Malware scan: pass → unverified

### 2026-10-31 (score 76, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-30 (score 61, −15)

- [security regression] Malware scan: pass → unverified

### 2026-10-27 (score 76, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-26 (score 61)

First indexed and scored.

## MCP tools (8)

### `remember` (~259 tokens)

Store a memory. Use for facts, decisions, preferences, or lessons learned during coding.

Args:
    content: The memory to store
    memory_type: One of: fact, decision, preference, lesson, summary, context
    project: Project name (optional, for scoping)
    related_to: IDs of related memories (optional)
    ttl: Expire after duration (e.g. "30d", "12h", "1w3d") for time-sensitive facts
    expires_at: ISO-8601 datetime to expire at (mutually exclusive with ttl)
    supersedes: ID of a memory this one replaces; the old memory is tombstoned (restorable for 7 days)
    check_conflicts: Dry-run LLM conflict detection only; returns candidates, writes nothing
    auto_supersede: If a high-confidence conflict (>=0.85) is found, supersede that memory

Input parameters:

- `auto_supersede` (boolean)
- `check_conflicts` (boolean)
- `content` (string, required)
- `expires_at`
- `memory_type` (string)
- `project`
- `related_to`
- `supersedes`
- `ttl`

Output parameters:

- `result` (string)

### `edit` (~198 tokens)

Edit a memory in place. Preserves id, created_at, and source. Bumps updated_at.

Pass only the fields you want to change. clear_project / clear_expiry explicitly null those fields.

Args:
    id: Memory ID to edit
    content: New content (optional)
    memory_type: New type: fact, decision, preference, lesson, summary, context (optional)
    project: New project name (optional)
    clear_project: Set project to null
    ttl: Reset expiry to a duration (e.g. "30d")
    expires_at: Reset expiry to an ISO-8601 datetime
    clear_expiry: Remove expiry: make memory permanent

Input parameters:

- `clear_expiry` (boolean)
- `clear_project` (boolean)
- `content`
- `expires_at`
- `id` (string, required)
- `memory_type`
- `project`
- `ttl`

Output parameters:

- `result` (string)

### `recall` (~127 tokens)

Search memories by natural language query.

Args:
    query: What to search for
    project: Scope to a specific project (optional)
    memory_type: Filter by type: fact, decision, preference, lesson (optional)
    since: Only memories created at or after this point: ISO date (2026-06-01) or duration (7d, 1w3d)
    limit: Max results (default 10)

Input parameters:

- `limit` (integer)
- `memory_type`
- `project`
- `query` (string, required)
- `since`

Output parameters:

- `result` (string)

### `recall_index` (~95 tokens)

Cheap first pass: returns IDs and snippets only. Use recall_full to fetch full content for relevant IDs.

Args:
    query: What to search for
    project: Scope to a specific project (optional)
    memory_type: Filter by type (optional)
    limit: Max results (default 20)

Input parameters:

- `limit` (integer)
- `memory_type`
- `project`
- `query` (string, required)

Output parameters:

- `result` (string)

### `recall_full` (~46 tokens)

Fetch full content for a batch of memory IDs returned by recall_index. Always batch.

Args:
    ids: Memory IDs from a prior recall_index call

Input parameters:

- `ids` (array, required)

Output parameters:

- `result` (string)

### `forget` (~31 tokens)

Delete a memory by ID.

Args:
    id: The memory ID to delete

Input parameters:

- `id` (string, required)

Output parameters:

- `result` (string)

### `consolidate` (~96 tokens)

Store learnings from a coding session. Call at session end to persist what you learned.

Args:
    session_summary: Brief summary of what was done in the session (stored as a 'summary' memory)
    facts: List of facts, decisions, preferences, or lessons learned
    project: Project name (optional, for scoping)

Input parameters:

- `facts` (array, required)
- `project`
- `session_summary` (string, required)

Output parameters:

- `result` (string)

### `context` (~62 tokens)

Get the top-N most recent memories for a project. Useful for session-start priming.

Args:
    project: Scope to a specific project (optional)
    limit: Max memories to return (default 20)

Input parameters:

- `limit` (integer)
- `project`

Output parameters:

- `result` (string)

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/ai-trags-poppy/poppy-memory#diagnostics

## Score history

- 2026-11-05: 81
- 2026-11-04: 61
- 2026-11-03: 61
- 2026-11-02: 76
- 2026-11-01: 76
- 2026-10-31: 76
- 2026-10-30: 61
- 2026-10-29: 76
- 2026-10-28: 76
- 2026-10-27: 76
- 2026-10-26: 61

## Common questions

### What is the ai.trags/poppy MCP server?

ai.trags/poppy is an MCP server listed in the public MCP registry as ai.trags/poppy. Developer memory CLI and MCP server. Remember decisions, preferences, and lessons locally. This page covers its PyPI package (poppy-memory).

### Is the ai.trags/poppy MCP server safe to use?

ai.trags/poppy scores 81 out of 100 on VerifyMCP. Its build provenance is signed and verified. 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 ai.trags/poppy MCP server expose?

ai.trags/poppy exposes 8 tools: remember, edit, recall, recall_index, recall_full, and 3 more. Their descriptions and schemas cost roughly 914 tokens of context every time the server is loaded.

### Is the ai.trags/poppy MCP server still maintained?

ai.trags/poppy is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the ai.trags/poppy MCP server under?

ai.trags/poppy declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/poppy-memory/
- Socket report: https://socket.dev/pypi/package/poppy-memory
- Repository: https://github.com/trags-garden/poppy
- Website: https://trags.ai/poppy
- Changelog RSS feed: https://verifymcp.io/servers/ai-trags-poppy/poppy-memory.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-trags-poppy/poppy-memory.json
- HTML version of this page: https://verifymcp.io/servers/ai-trags-poppy/poppy-memory