dsh-darwin
Dual-plugin self-evolution architecture for DeepSeek Harness featuring session log mining, tiered skill synthesis, eval gating, and rollback.
Install
$ dsh plugin --profile <你的profile> add dsh-darwin-sentinelPlugin 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
- Mechanical session log mining with fingerprint deduplication and severity scoring (`darwin_scan`, `darwin_report`)
- Tiered synthesis of configuration patches and Markdown skills (`darwin_forge`)
- Gated evaluation with hold-out canary checks and cost fitness enforcement
- Hot-reloading skill promotion into project-level `.dsh/skills/` (`darwin_promote`)
- Snapshot-based deterministic rollback (`darwin_rollback`)
Useful For
- Diagnosing agent failure modes and repetitive tool errors across DeepSeek Harness sessions
- Generating and iteratively improving agent skills based on historical execution tickets
- Automating continuous self-improvement workflows with deterministic safety rollbacks
Who It Fits
- DeepSeek Harness plugin developers
- AI agent engineers managing multi-turn DSH workflows
- Teams evaluating autonomous agent self-refinement loops
Documented Limitations
- Code-level synthesis (Creator mode) is disabled by default for sandbox security
- Evaluation gate A/B fork trials require upstream DSH dynamic Cordis runner and session fork support
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, 4 development, 0 peer, and 0 optional dependencies.
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
- que3sui/dsh-darwin
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin Bundle
- Source checked
- ecd03cb · 2026-10-25
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
- 1GitHub stars
- Forks
- 0
- Open issues
- 7
- Last commit
- 2026-10-19
- Last release
- 2026-09-02
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Comprehensive long-term memory, self-evolution, skills, task management, and multi-agent dispatch bundle for DeepSeek Harness.
View plugin →Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Second-model AI auto-review for DeepSeek Harness approval requests with fail-closed security.
View plugin →Native SQLite-backed task management and project board bundle with Web UI and Agent workflow tools for DeepSeek Harness.
View plugin →