inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail
io.github.DanceNitra/inspeximus
About inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail
inspeximus: agent memory with provenance, GDPR erasure and a tamper-evident audit trail is an MCP server published by the GitHub account DanceNitra. Self-correcting agent memory: recall, supersede, revert, forget, GDPR erasure, audit receipts.
It runs on your own machine: your AI client starts it from the PyPI package inspeximus at version 3.23.1.
It asks for no keys or passwords.
Version 3.23.1 was published on 26 October 2026, and the source code is public, so you can read it before installing.
Written from the registry entry. 9 automated checks on the registry entry and the package code: 5 pass, 2 warn, 0 fail, 2 info. How we check.
What it asks for
| Variable | Required | Secret | Description |
|---|---|---|---|
INSPEXIMUS_PATH | No | No | Path to the JSON memory file the server persists to (defaults to ./inspeximus_memory.json). |
INSPEXIMUS_ECHO_GUARD | No | No | 1 (default) blocks a restated retired value from resurrecting a corrected fact; 0 = legacy keyed supersession. |
INSPEXIMUS_EMBED_URL | No | No | Optional OpenAI-compatible /embeddings endpoint for SEMANTIC recall (e.g. http://localhost:11434/v1/embeddings for local Ollama). Unset = deterministic lexical recall. |
INSPEXIMUS_EMBED_MODEL | No | No | Embedding model id used with INSPEXIMUS_EMBED_URL (default: text-embedding-3-small; for Ollama use e.g. nomic-embed-text). |
Checks
| Check | Result | Detail |
|---|---|---|
| Source repository linked | Pass | https://github.com/DanceNitra/inspeximus |
| Version pinned | Pass | Every package has an explicit version |
| Secrets requested | Info | Asks for no secrets |
| Known package registry | Pass | All packages come from npm, PyPI, OCI, NuGet or MCPB |
| Namespace ownership | Info | Name verified via GitHub account DanceNitra |
| Registry text aimed at the AI | Pass | Description and settings contain no hidden characters or instructions to the AI |
| Install scripts | Pass | inspeximus 3.23.1 has no setup.py; nothing runs at install |
| Code red flags | Warn | Reads SSH or cloud credentials (inspeximus-3.23.1/inspeximus/_secrets.py:171: "id_rsa") |
| Text aimed at the AI | Warn | Tells the AI to hide something from the user (inspeximus-3.23.1/inspeximus/core.py: "…ect between the verb and the audience: 'do not mention this note to the user' re.compile(r"\b(do no…"). Read the context: a security tool may quote these on purpose. |
Popularity & health: 51 / 100
Worked out from public numbers and our own checks, not from user reviews or ratings. It says how used and how looked-after a server is, not whether it is safe.
1 forks, 27 merged pull requests. Sources: GitHub, npm, PyPI (pypistats.org) and Hacker News (Algolia), fetched 2026-10-31.
What people say
No Hacker News threads link to its source code.
Look further: search Reddit, search Hacker News, read its GitHub issues. These are not reviews we checked.
Facts
| Registry name | io.github.DanceNitra/inspeximus |
|---|---|
| Publisher | GitHub account DanceNitra |
| How the name is verified | Name verified via GitHub account DanceNitra. The registry ties the name prefix to that account or domain; it does not check the code. |
| Type | Local: runs on your machine from a package |
| Packages | pypi inspeximus 3.23.1 (stdio) |
| Version | 3.23.1 |
| Published | |
| Last updated | |
| Source repository | https://github.com/DanceNitra/inspeximus |
| Website | https://pypi.org/project/inspeximus/ |
| Category | Productivity & notes (assigned by keyword rule, not by hand) |
Related servers in Productivity & notes
- Attestari7/8Auditable memory for AI agents: provable deletion, tamper-evident…
- Lians Agent Memory7/7Bitemporal memory for AI agents with point-in-time recall and…
- cuba-memorys6/7Persistent memory MCP server. 28 tools, BM25+MMR+OOD retrieval,…
- mcp7/7Private memory layer for AI assistants. Local-first: remember,…
- Mindtrail7/7Persistent, local-first memory for AI coding agents. Remember,…
- Engram6/7Cognitive memory for coding agents: hybrid recall, remember/forget,…