dsh-layered-memory
Cross-session layered long-term memory for DeepSeek Harness with structured identity, evidence tracing, and rollback history.
Install
$ dsh plugin --profile web add github:DDDFXYqiming/dsh-layered-memoryPlugin 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
- Cross-session memory persistence across project-specific and global namespaces
- Dual storage hierarchy supporting structured facts in markdown headings and procedures in SOP files
- Compact L1 context injection into system prompts with character budget bounds and folding
- Structured addressing via namespace, kind, and key tuples rather than display titles
- Evidence inspection and full version history tracking with snapshot rollback capabilities
- Granular access control supporting read-only sessions and subagent isolation
- Optional background maintenance subagents for reflection and memory candidate grooming
Useful For
- Persisting project facts, architectural decisions, and build commands across separate agent sessions
- Maintaining workspace-specific standard operating procedures and runbooks that models can reference on demand
- Rolling back accidental or inaccurate memory updates to previous verified snapshots
- Providing read-only historical context to child or delegated subagents
Who It Fits
- DeepSeek Harness users working on recurring, long-running codebase projects
- Developers seeking deterministic, verifiable cross-session agent memory and SOP retrieval
- Agent workflow engineers designing multi-agent setups requiring shared workspace knowledge
Documented Limitations
- When memory indices exceed character budgets, entries are folded and descriptions are trimmed until explicitly read via tools
- Automatic reflection or maintenance can trigger subagents that consume additional model tokens per run
- Automatic namespace assignment relies on canonical root paths or workspace identifiers, requiring explicit migration if paths change
- Memory tools and access are only exposed after explicit memory activation in a session
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 0 runtime, 7 development, 4 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.
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
- DDDFXYqiming/dsh-layered-memory
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin Bundle
- Source checked
- e384c70 · 2026-10-19
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
- 1
- Open issues
- 0
- Last commit
- 2026-10-19
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
LLM-supervised persistent memory and graph-based recall integration bundle for DeepSeek Harness.
View plugin →Persistent semantic memory layer for DeepSeek Harness backed by Markdown and Milvus.
View plugin →Composable, view-based layered memory bundle for DeepSeek Harness with pluggable providers and strategies.
View plugin →Local cross-agent memory and session history recall integration for DeepSeek Harness.
View plugin →