dsh-evidence-ledger
Run the following command in DeepSeek Harness:
dsh plugin install Freakz2z/dsh-evidence-ledger
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install Freakz2z/dsh-evidence-ledger from your DeepSeek Harness workspace root to install this plugin; the source is available at https://github.com/Freakz2z/dsh-evidence-ledger .
About this plugin
In multi-step DSH agent sessions, observed test results, decisions made, and failure paths tend to vanish once context is trimmed or the conversation resets. Without a structured record, debugging and reproducibility become nearly impossible. dsh-evidence-ledger pins these critical facts into an append-only JSONL file inside the workspace, creating a local evidence ledger that never phones home and never deletes prior entries.
The evidence_ledger tool classifies entries by kind (fact, test, decision, failure, note) and status (observed, verified, rejected, pending). list applies exact filters on kind, status, and tag; query performs case-insensitive full-text search across claims, evidence, sources, and tags; summary returns total matches, aggregated counts by kind and status, and the most recent matching record. By default the ledger lives at .dsh/evidence-ledger.jsonl, with paths restricted to workspace-relative locations, rejecting absolute and traversal paths.
This plugin is well suited to developers running multi-step reasoning, automated testing, or code-review workflows where a traceable evidence trail matters. It does not judge whether a claim is true; it simply preserves raw records and provides searchable source pointers, so every conclusion can be audited later.
Use Cases
- Preserving test results and decisions across multi-step reasoning
- Tracing failure paths and verification status during code review
- Retaining critical facts that would be lost to context trimming in long sessions
Best For
- Developers running automated tests with DSH agents
- Research or engineering teams that need an auditable evidence trail
- Architects building multi-step workflows where traceability matters
Related Plugins
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.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.