dsh-memory
Cairn is a long-term cross-session memory profile bundle for DeepSeek Harness with BM25 retrieval and human review.
Install
No clear install command was found in the current repository evidence.
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.
Key Capabilities
- Persistent KV memory store with two-tier lifecycle decay and audit logging
- Three-layer scoping across global, project-specific, and user profile boundaries
- Dependency-free Okapi BM25 relevance search supporting Latin words and CJK n-grams
- Ten agent-facing memory and identity tools including memory_search, memory_add, and identity_update
- Automated learning from explicit intent, corrections, and resolved tool failure streaks
- Digest-first and step-level auto recall message injection preserving KV-cache stability
- Prompt-level security scanning and redaction for secrets and prompt injections
- Comprehensive UI settings manager with Health Overview, Review proposal queue, and Entry management
Useful For
- Retaining coding preferences, conventions, and architectural decisions across DSH sessions
- Tracking resolved pitfalls and failure patterns so the agent does not repeat mistakes
- Maintaining persistent user profiles and agent character identity without modifying repository files
- Governing agent learning through human confirmation queues before memories are committed
Who It Fits
- Developers utilizing DeepSeek Harness who require cross-session context continuity
- Engineers managing complex multi-session projects needing automatic pitfall tracking
- Users wanting customized and persistent AI assistant personas and preferences
Documented Limitations
- Retrieval relies on BM25 lexical token matching rather than vector embeddings or semantic search
- Requires DeepSeek Harness alpha channel (0.1.2-alpha.x) and is incompatible with the rc (next) line
- Headless DSH profiles require manual storage layer configuration
- Mid-session extractions do not enter the system prompt until compaction or a new session boundary
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 3 runtime, 37 development, 13 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
package.json declares a prepare lifecycle script that may run during relevant package installation workflows.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- chenhw7/dsh-memory
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin Bundle
- Source checked
- 401a729 · 2026-10-23
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
- 3GitHub stars
- Forks
- 3
- Open issues
- 1
- Last commit
- 2026-09-10
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Knowledge graph memory and context engine for DeepSeek Harness with cross-session recall.
View plugin →Persistent semantic memory layer for DeepSeek Harness backed by Markdown and Milvus.
View plugin →Session lifecycle manager for DeepSeek Harness with search, restore, idle cleanup, and diagnosis.
View plugin →Composable, view-based layered memory bundle for DeepSeek Harness with pluggable providers and strategies.
View plugin →