dsh-ca-ref
Run the following command in DeepSeek Harness:
dsh plugin install Ansonfishing/dsh-ca-ref
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install Ansonfishing/dsh-ca-ref in DeepSeek Harness to install this plugin; the source repository is at https://github.com/Ansonfishing/dsh-ca-ref
About this plugin
Auditing a Clean Architecture, DDD, or hexagonal-layered codebase usually means wading through scattered reference repos—architecture tests, design docs, assertion suites—every single time. dsh-ca-ref eliminates that repetition by pre-indexing eight pinned reference repositories (Go, Java, C#, Python) into a local SQLite FTS5 database. A single ca_ref_baseline call returns the full language-specific baseline: summary cards, a rule checklist with assertion templates and source citations, and eight red-flag rules—ready before review, logged after.
Four Agent tools (ca_ref_search, ca_ref_baseline, ca_ref_record, ca_ref_status) plus a /ca-ref command channel cover the entire workflow: full-text search across assertions, docs, and structure; one-shot baseline retrieval; a post-review ledger capturing project, language, matched rules, and evidence; and library health checks including pinned-SHA verification. Corpus integrity is guarded by a daily read-only git rev-parse HEAD check—drift raises a warning badge in the panel but never triggers an automatic re-clone, keeping every audit trail reproducible. A read-only observation panel auto-refreshes every 30 seconds, surfacing rule coverage, search history, review ledger, and hit statistics in one view.
Built for developers, tech leads, and code-review owners who regularly audit multi-language Clean Architecture projects and want to turn "dig through docs" into "query an index."
Screenshots
Use Cases
- Audit whether Go, Java, C#, or Python projects follow Clean Architecture layering conventions
- Retrieve assertion templates, architecture rules, and red-flag items from pinned reference repos before a review session
- Log matched rules and evidence after review to build hit statistics that feed back into corpus curation
Best For
- Engineers accountable for multi-language code review and architecture compliance
- Tech leads driving DDD or hexagonal adoption across teams
- Architecture governance and QA roles that need auditable review records
Related Plugins
Memory layer for coding agents that indexes local session history and auto-recalls relevant context before edits, commands, and failures, with no manual search needed.
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.