Overview

Provide a simple CLI-based long‑term memory layer for agents by adding, searching, and chatting over a user’s stored content via the SuperMemory API.

Key Advantages

1.Very simple interface: three clear operations (add, search, chat) mapped to straightforward CLI commands and bash scripts.

2.Enables persistent memory across sessions, useful for recalling prior context and user data.

3.Supports semantic search and conversational querying over the stored knowledge base, not just keyword lookup.

4.Integrates cleanly into shell-based workflows or agent toolchains via the provided bash scripts.

Use Cases

Personal knowledge management for non-sensitive notes, articles, and project summaries.
Agent long‑term context store for things like user preferences, recurring tasks, or project history (excluding secrets).
Semantic search over accumulated research notes or documentation stored over time.
Conversational review or summarization of what has been learned about a topic (e.g., “Summarize what I’ve learned about AI”).
Bookmarking and later retrieval of links and short snippets that are safe to store in a third‑party service.

Evaluation Scores

6.5

/ 10

Reliability

6.5

Functionality

7.5

Usability

8.0

Safety

3.0

Performance

7.0

Compatibility

7.5

Based on 1 evaluation · Latest: 3/19/2026

Download Trend

Loading...

Evaluation History (1)

6.5/103/19/2026linux-arm64

▼

OS: linux-arm64LLM: z-ai/glm-5

**Judgment:** A useful, minimalistic long‑term memory/PKM skill that wraps the SuperMemory API via CLI, suitable for non‑sensitive knowledge storage and retrieval. Functionally solid for “add/search/chat” over a knowledge base, but with significant security and privacy caveats. **What it does well** - Provides three core operations: `add` (store text with optional description), `search` (find relevant memories), and `chat` (conversational querying over stored content). - Simple to integrate in agent workflows via bash scripts (`add-memory.sh`, `search.sh`, `chat.sh`). - Good example prompts illustrate how to use it for recalling notes, links, and topic summaries. **Key risks & concerns** - **Sensitive data leakage:** The examples explicitly encourage storing secrets (e.g., “Remember that my API key is xyz”). Any such data is sent to and stored by the SuperMemory service, which is inappropriate for credentials or highly sensitive personal information. - **Third‑party data storage:** All stored content resides in an external API with no documented guarantees here about encryption, retention, or access control. This is problematic for regulated or confidential data. - **Security hygiene:** The documentation shows an API key being set in an environment variable and even includes what looks like a concrete key in the README text. Even if this is a placeholder, it models insecure practices and raises concerns about key handling. - **Operational dependency:** Reliability depends on the external SuperMemory API and the presence/configuration of `SUPERMEMORY_API_KEY`; failure modes and error handling are not described. **Recommended scenarios** - Use for **non‑sensitive** personal or project notes, research summaries, and general knowledge that can safely reside in a third‑party service. - Suitable as a **long‑term memory extension** for agents that need to recall user preferences, project summaries, or topic history that isn’t confidential. - Avoid for any secrets (API keys, passwords), proprietary data, or personally identifiable information that you wouldn’t store in a typical cloud PKM/notes app without strong guarantees. **Overall:** Technically useful and easy to adopt, but should be deployed with strict content restrictions and clear user guidance against storing secrets or sensitive information.

Supermemory