DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to give an agent memory that survives between sessions with Xerj

Store what an agent learns in a namespace and recall it by meaning in the next session.

Authorization notice. Xerj is dual-use. Only use it on software, sites and systems you own or have explicit permission to test.

Credentials. The --insecure start makes every caller a superuser who can read and delete every /_memory namespace, so do not store secrets in memory notes.

Who it is forA developer who restarts agent sessions on one project and wants the agent to keep notes on decisions and gotchas.

The problemEach new agent session starts from zero, so decisions and project quirks have to be re-explained every time.

The outcomeThe agent stores notes in a memory namespace and recalls relevant ones by meaning at the start of the next session.

What you need first

  • A Linux, macOS or Windows machine on x86-64 or arm64
  • A local folder of code, docs, logs or PDFs to index
  • Claude Code or Cursor if you want the MCP and skill setup from xerj init
  • A terminal with curl and a shell

Steps

  1. xerj --insecure --data-dir ~/xerj-data &
  2. Send a POST request to http://localhost:9200/_memory/notes with the note text as the body. (inferred, check the README)
  3. Query http://localhost:9200/_memory/notes with a question to recall notes by meaning. (inferred, check the README)
  4. Start the next session and ask the agent to recall notes on the project before working.

Prompt for your agent

Use XERJ's /_memory/notes API on http://localhost:9200 as long-term memory for <project>: store each decision you make, and recall relevant notes at the start of the session. Expected output: a short list of recalled notes with their dates. Verify that the notes returned match the stored text.

Caveat. The README names the /_memory/{namespace} path but does not give the HTTP methods or request bodies, so the request steps above are inferred and should be checked against Xerj.org/docs.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page