---
name: stijnman-semantic-memory-manager
description: "Stores and retrieves semantic memory with embeddings and tags. Use for: semantic memory, search memory, recall, store memory."
license: "MIT"
metadata:
  decimalai-skill-id: "1ebfe242-6328-48fe-9d13-3d735e5a1bd8"
  decimalai-version-id: "af99a228-9706-4f7e-bb0f-1885c94660d1"
  decimalai-version: "2"
  decimalai-body-sha256: "0968ad9069d5e0e6e9ef3758fa838fc8a453c8d413fb390b0b723f682c79afbd"
  decimalai-name: "stijnman/semantic-memory-manager"
  decimalai-format: "1"
  decimalai-invocation: "model"
  decimalai-category: "agent-tooling"
  decimalai-source: "https://app.decimal.ai/s/stijnman-semantic-memory-manager@2/SKILL.md"
---
# Semantic Memory Manager
## When to Use

- User says **semantic memory** or task matches this capability
- User says **search memory** or task matches this capability
- User says **recall** or task matches this capability
- User says **store memory** or task matches this capability

## Workflow

1. On store: chunk, tag, score initial trust 5.
2. On retrieve: query by semantic similarity + tags.
3. Run memory-sanitizer on all retrievals.
4. Prune entries older than 90d with trust < 3.

## Integrations

- `memory-sanitizer`
- `persistent-memory-bridge`
- `knowledge-graph-builder`

## Error Handling

| Failure | Response |
|---------|----------|
| No match | Broaden query; suggest manual tags. |

## Gotchas

- Central memory hub for messenger and healing skills.

## Example

**Input:** User request matching triggers above.
**Output:** Structured result per workflow with integrations invoked as needed.