dsh-deadend
Run the following command in DeepSeek Harness:
dsh plugin install liyixuan201211/dsh-deadend
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install liyixuan201211/dsh-deadend in DeepSeek Harness to install this plugin; the full source repository is at https://github.com/liyixuan201211/dsh-deadend .
About this plugin
Most agent memory systems store what worked — facts, preferences, embeddings — and almost nothing stores what failed. The same broken install, the same two-hour detour, gets paid again every session.
dsh-deadend records proven failures in a ledger that lives inside your repository. Every entry carries anchors — content hashes of the files that would have to change for the failure to stop being true. Anchors unchanged means the block holds; anchors changed means the entry auto-downgrades to a re-test state, because the code world it was written against no longer exists. Failure descriptions are normalised (stripping paths, line numbers, UUIDs, timestamps) so the same breakage is recognised across sessions and machines. Matching is layered: weak hits only inform, never block, so the gate never cries wolf.
Built for developers who lean on coding agents daily, teams that want a CI gate in one line, and anyone tired of an agent re-attempting a known dead end. The ledger is an append-only event log committed to the repo — reviewable in a pull request, shared across teammates and CI, concurrency-safe by design, and mergeable via content-derived identity with no de-duplication step.
Use Cases
- An agent re-attempts a known-broken install command in every new session
- A team shares dead-end knowledge that auto-expires when the code it targeted changes
- A CI pipeline blocks a fix that has already been proven not to work
Best For
- Developers who rely on coding agents daily
- DevOps teams that want a one-line CI gate
- Solo developers tired of agents repeating known failures
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.