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

How to use DSH Memento: what you can do with it

What is DSH Memento? A DeepSeek Harness plugin that gives agent sessions bounded, approval-gated, auditable cross-session memory stored in a local SQLite database.

How do I use it? Run dsh plugin --profile web add dsh-memento to install the plugin; then run dsh --profile web --dump-config | grep -A3 'id: memento' to confirm the memento row is present; run /memory add and store one convention, such as the test command, in the agent track workspace layer. Full steps below.

Trended earlier

View repo on GitHub

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer

What do you want to do?

4 use cases

  • Keep workspace conventions across agent sessions

    Store project facts once so each new DeepSeek Harness session starts with them in the system prompt.

    View prompt
    In the dsh-memento repo, store the test command for <project> in the agent track workspace layer with /memory add, then run /memory list. Expected output: the new entry listed under agent and workspace. Verify that the memory_stats total increased by one.

  • Review every memory write before it is saved

    Keep writePolicy at ask so each add, replace, or remove waits for your approval.

    View prompt
    In the dsh-memento repo, confirm writePolicy is ask, trigger one memory write from <session>, deny it, then run /memory audit. Expected output: a *-denied audit row for that write. Verify that /memory list does not show the denied entry.

  • Audit what the model saw and what it wrote

    Rebuild the memory write history from the plugin audit table and the session log.

    View prompt
    Using the dsh-memento repo on profile <profile>, run /memory audit and list each write with its outcome. Expected output: rows from the plugin audit table, plus the session-log gap line if it is shown. Verify that each write has a matching approval/asked and approval/decided pair in the session log.

  • Find earlier session discussions with memory_recall

    Search a bounded set of past session matches so a new session can pick up an earlier decision.

    View prompt
    In the dsh-memento repo, call memory_recall with the query "<topic>" and list the matched snippets. Expected output: bounded memory matches and recent session-history matches. Verify that no session returns more snippets than the recall.snippetCap setting.

Quick start with DSH Memento

  1. Check you have what it needs: Node ^22.19.0 or >=24.0.0; DeepSeek Harness with the dsh CLI installed; A dsh profile, such as the web profile used in the examples.
  2. Open PerryLink/dsh-memento on GitHub and read the README and license.
  3. Run dsh plugin --profile web add dsh-memento to install the plugin.
  4. Run dsh --profile web --dump-config | grep -A3 'id: memento' to confirm the memento row is present.
  5. Run /memory add and store one convention, such as the test command, in the agent track workspace layer. (inferred, check the README)
  6. Start a new session and check that the stored convention is in the injected snapshot.

These steps come from our top use case, Keep workspace conventions across agent sessions, which also has a copy-paste agent prompt.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for PerryLink/dsh-memento

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

DSH Memento alternatives · Best AI repos for Software and DevOps

About this repo

  • Who it is forDevelopers on DeepSeek Harness who repeat the same project facts across sessions and want control over what the model remembers.
  • Why it is hot141 stars, launched 74 days ago
  • LanguageJavaScript
  • Sourcegithub.com/PerryLink/dsh-memento · Back to Hot repos

Requirements

  • Node ^22.19.0 or >=24.0.0
  • DeepSeek Harness with the dsh CLI installed
  • A dsh profile, such as the web profile used in the examples

Add this badge to your README

Maintain DSH Memento? Show readers where to find these use cases. Use cases on Deduped

Markdown 

[![Use cases on Deduped](https://dedupedai.com/badge/PerryLink/dsh-memento.svg)](https://dedupedai.com/use-cases/perrylink-dsh-memento/?utm_source=github&utm_medium=badge&utm_campaign=PerryLink-dsh-memento)

HTML 

<a href="https://dedupedai.com/use-cases/perrylink-dsh-memento/?utm_source=github&amp;utm_medium=badge&amp;utm_campaign=PerryLink-dsh-memento"><img src="https://dedupedai.com/badge/PerryLink/dsh-memento.svg" alt="Use cases on Deduped" height="20"></a>

A static image, no tracking script. About the badge

Questions about DSH Memento

What is DSH Memento?

A DeepSeek Harness plugin that gives agent sessions bounded, approval-gated, auditable cross-session memory stored in a local SQLite database. It is an open-source project on GitHub: https://github.com/PerryLink/dsh-memento.

How do I use DSH Memento?

Run dsh plugin --profile web add dsh-memento to install the plugin; then run dsh --profile web --dump-config | grep -A3 'id: memento' to confirm the memento row is present; run /memory add and store one convention, such as the test command, in the agent track workspace layer. The use cases on this page give the full steps and a copy-paste agent prompt for each job.

Who is DSH Memento for?

Developers on DeepSeek Harness who repeat the same project facts across sessions and want control over what the model remembers.

What can I do with DSH Memento?

4 use cases so far, for example: Keep workspace conventions across agent sessions; Review every memory write before it is saved; Audit what the model saw and what it wrote. Each has steps and a copy-paste agent prompt.

What do I need to run DSH Memento?

Node ^22.19.0 or >=24.0.0; DeepSeek Harness with the dsh CLI installed; A dsh profile, such as the web profile used in the examples.

What are alternatives to DSH Memento?

Other hot AI repos in Software and DevOps: Deep Infer, Harness2, Orca, Opencode, Graphify.

Is DSH Memento safe to run?

Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.