PensyveBLOCK
Universal memory runtime for AI agents
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://github.com/major7apps/pensyve/actions/workflows/ci.yml) [](LICENSE) [](https://www.python.org/downloads/) [](https://www.rust-lang.org/)
Pensyve is an open-source runtime for persistent AI agent memory. It stores facts, conversations, observations, and action outcomes so an agent can retrieve them in later sessions. You can use it through Python, a Model Context Protocol (MCP) server, a command-line tool, or a REST API.
The Rust engine uses SQLite for local storage and runs embedding models locally to search by meaning. TypeScript and Go clients connect to a gateway you run yourself. Local use does not require a Pensyve account or API key. Your application or client integration calls Pensyve to save and retrieve memories.
Choose a setup
Project status
Pensyve Cloud closed on October 1, 2026. The open-source project continues in this repository under the Apache 2.0 license, with the engine, SDKs, integrations, and documentation available here.
Pensyve is in maintenance mode. Releases cover security fixes and dependency updates, with no new features planned. See the [maintenance policy](MAINTENANCE.m
41f7cf64d0d9OBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add pensyve-mcp:5.0.0 -- docker run -i --rm ghcr.io/major7apps/pensyve-mcp:5.0.0:None
Exposed tools (8)
7 read · 0 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
memory_forget | destructive | Delete all memories. Use only when explicitly asked. |
memory_get | read | Get all stored memories for the current entity. |
memory_recall | read | Search Pensyve memory for facts, preferences, and context from prior sessions. |
memory_status | read | Show Pensyve connection status, memory counts, and account info. |
memory_store | read | Store a fact in persistent memory. Use present tense. |
pensyve | read | Search Pensyve memory (default) or show status with |
pensyve-auto-capture | read | Stores the last exchange after each agent response. |
pensyve-auto-recall | read | Injects recalled Pensyve memories before prompt build. |
Trust audit
BLOCKgrade F · trust 48/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | FAIL |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (9 observation(s))
- Network
- declared (11 observation(s))
- Shell
- declared (5 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
const TEST_ED25519_PRIVATE_PEM: &str = "-----BEGIN PRIVATE KEY-----\n\
exec(
exec(&admin, format!("CREATE DATABASE \"{database}\"")).await;exec(pool, "CREATE EXTENSION IF NOT EXISTS vector").await;
exec(
exec(
let url = "redis://sensitive-user:[email protected]:6379/0";
If no relevant memories are found, or all scores are below 0.3, proceed without enrichment. Do not inform the user that enrichment was attempted. Do not show "no memories found" messages.
p.check("http://127.0.0.1:8888/v1/chat/completions")std::env::set_var("HF_ENDPOINT", "http://127.0.0.1:9");url: "http://127.0.0.1:9".to_string(),
PENSYVE_TEST_DATABASE_URL: postgres://postgres:postgres@localhost:5432/postgres
memory_forget
.vscodeignore
} from "../../shared/pensyve-client";
} from "../../shared/pensyve-client";
} from "../../shared/pensyve-client";
let lock = include_str!("../../Cargo.lock");await run(`http://127.0.0.1:${port}`, seen);let allowed = "http://127.0.0.1:1/v1"; // port 1 is privileged + unbound
typescript, vitest
typescript, vitest
@opencode-ai/plugin, typescript, vitest
vitest
@types/node, @types/vscode, @vscode/vsce, typescript
Gates applied: critical_finding, instruction_override, no_behavioural_pass.
41f7cf64d0d9full audit observations/trust-audit/mcp-server/major7apps__pensyve.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 41f7cf64d0d9 | BLOCK | F | 48 | first audit |
Questions
What is the Pensyve MCP server?
Universal memory runtime for AI agents
What tools does Pensyve expose?
8 in total: 7 read-only, 0 that write, and 1 that can delete or overwrite (memory_forget). Every one is listed on this page with its risk.
Is Pensyve safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (48/100) and found 8 critical or high issues in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Pensyve need?
It reads PENSYVE_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Pensyve run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as @pensyve/sdk at 5.0.0.
How current is this page?
The grade is for one exact copy of the source (41f7cf64d0d9), read on 2026-10-07. The repository is watched and re-audited when it changes.