DeepSeek Harness Plugins
DeepSeek Harness PluginIndexed

causal-memory

Causal memory layer and tool suite for AI agents modeling decision-outcome relationships and inhibition.

Developer ToolsSecurity & PolicySkills & WorkflowsRemote Execution
52GitHub Stars3ForksUpdated2026-08-29

Install

$ One-line install: dsh plugin --profile web add "$PWD/dsh-plugin" (needs the

Plugin Overview & Capabilities

AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.

source-grounded
causal-memory is an agent memory system that stores facts, temporal states, and decision-to-outcome causal graphs in an external SQLite database resistant to context compaction. It includes a native DeepSeek Harness integration (dsh-plugin) providing direct access to causal memory tools, forward simulation, counterfactual queries, and memory consolidation.

Key Capabilities

  • Records and retrieves typed causal edges (caused, enabled, prevented)
  • Provides inhibitory spreading activation (GABA analogue) for danger and failure warnings
  • Offers forward intervention simulation and counterfactual queries before taking actions
  • Integrates 14+ tools directly into DeepSeek Harness ctx.tools via JSON-RPC stdio
  • Features hippocampal-style offline consolidation (SWR), Hebbian reinforcement, and Q-value dynamics

Useful For

  • Preventing autonomous agents from repeating known mistakes across long sessions and context compactions
  • Analyzing root causes of failed operations through multi-hop backward causal graph traversal
  • Evaluating alternative decision paths prior to high-risk tool or shell executions

Who It Fits

  • Autonomous agent developers
  • DeepSeek Harness power users requiring persistent long-term memory
  • AI workflow designers managing complex, multi-step actions

Documented Limitations

  • Requires the underlying causal-memory binary (via Cargo build or pip) to communicate over stdio
  • Cross-task analogy transfer (meta-edge learning) remains an active development area with lower baseline accuracy
  • HTTP transport mode endpoints are unauthenticated and should not be exposed to the public internet

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

License Declared

GitHub reports the repository license as Apache-2.0.

info
Package Manifest Not Found

No root package.json was captured in the latest GitHub snapshot.

info
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
JingxuanC/causal-memory
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
c162eef · 2026-08-31

This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.

Repository Activity

GitHub stars
52GitHub stars
Forks
3
Open issues
10
Last commit
2026-08-29
Last release
2026-08-29
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app