Repo · Context & memory
deepseek-harness-jev-pre-compaction
A pre-compaction advisor for DeepSeek Harness.
Open github.com ↗From the repository
A pre-compaction advisor for DeepSeek Harness. Runs before the standard `compaction-basic` backend, using TypeSafe JEV to safely prune low-value tool results from model context. Original session events stay in the append-only log; only the model-visible view is replaced with compact markers or archive pointers to reduce context bloat.
How builders describe it
A pre-compaction advisor for DeepSeek Harness. Runs before the standard `compaction-basic` backend, using TypeSafe JEV to safely prune low-value tool results from model context. Original session events stay in the append-only log; only the model-visible view is replaced with compact markers or…
Pre-compaction advisor for DeepSeek Harness that asks Jev which tool results can be pruned in the 70-80% lookahead band, keeping the log intact and leaving real compaction to the backend.
The decision Jev makes
Keep or discard a tool result, or answer whether a stopping condition is met.
Where it fits
Agents drown in their own context. These projects use Jev to decide what to keep, what to drop and when a loop is actually finished. All 117 context & memory projects →
Related projects
Context & memoryfast-compact/fc for Claude Code: TypeSafe Jev picks which old tool outputs are still needed and keeps them whole, other outputs are cut to head/tail with every…GitHub · ★ 6 · NodarDavituri
Context & memorypi-context-decisionsPi extension that uses decision models (TypeSafe Jev, OpenRouter decision models, OpenAI Luna) to filter which skill advertisements enter context for…GitHub · championswimmer
Context & memoryOpen Memory (pi)Local-first SQLite memory for Pi agents with hybrid retrieval and a hard token budget; an optional TypeSafe Jev judge decides which sentences are…GitHub · ★ 1 · codersdfs
Context & memoryJev Gate (royalskynet)MCP server and Claude Code/Codex plugin that uses Jev to keep only the parts of logs and tool output that matter, shorten long conversations, and…GitHub · royalskynet
Context & memoryOpenVikingVolcengine's agent context database that unifies memory, knowledge RAG and skills under a viking:// virtual filesystem; it ships a Jev rerank client…GitHub · ★ 40k
Context & memoryMinutesAn MIT-licensed, local-first conversation memory app that transcribes meetings, calls and voice memos on-device into Markdown you own.GitHub · ★ 1.5k