jevaction
A lightweight agent loop with Jev decisions and persistent reusable actions
26 results
A lightweight agent loop with Jev decisions and persistent reusable actions
Laya typed decisions (noul / choice / score) as a first-class Cordis service and model-visible tools for DeepSeek Harness.
Verdict-based compaction engine for DeepSeek Harness: replaces the lossy compaction summary with fast per-call keep/truncate/drop decisions, keeping everything else verbatim (port of tamaratran/fast-jev-compaction)
Retrieved ≠ injected: CBDC-gated memory for DeepSeek Harness — decides how memory is USED (use/verify/ignore decisions, feedback learning, full audit); local SQLite + FTS5, bounded, no extra model call
Work Continuity for DSH — goal/decisions/checkpoints/next-steps as explicit WorkState, decoupled from generic memory
Observational memory for DeepSeek Harness: background observers distill session work into observations and durable reflections, so long sessions survive compaction with their decisions intact.
DeepSeek Harness plugin: cross-session memory for the model. Full-text search all past sessions (SQLite FTS5 via ctx.sessionQuery) and bring the strongest matching excerpts back into the current context — recall earlier decisions, commands, and mistakes.
Track human decisions for related DeepSeek Harness conversation branches.
Focus board for DeepSeek Harness agents: durable, model-maintained notes in the session workspace that pin the objective, constraints, and decisions across compaction and sessions — with structured entry tags (kind/status), a grouped summary view, automatic context injection, an archive on clear, and an optional read-only web panel
Auto-capture key decisions from dsh sessions into a versionable DECISIONS.md and inject them into future sessions
Per-project memory for DeepSeek Harness: auto-creates memory/PROJECT.md, DECISIONS.md and SESSIONS.md per project, loads them into every session, and records durable outcomes by classification.
DeepSeek Harness WSL plugin: call TypeSafe Jev / OpenRouter System One for noul, choice, and score decisions (jev_ask / jev_check / jev_rank).
Durable delayed decisions that wake cold DeepSeek Harness sessions with bounded authorization
Deterministic regression tests for DSH policy decisions, evaluated without invoking the protected tool body.
Content-addressed proof that recorded DSH authorization decisions and effects agree
Policy-only multi-source quota router for DeepSeek Harness: deterministic task profiles, ordered candidate chains, health-aware fallback, and observable decisions.
📌 Sticky-note settings tab for DeepSeek Harness — agent notices & pending decisions visible in the Web GUI (Tailscale-friendly). Zero deps, read-only.
Persistent, searchable, per-project memory for the DeepSeek Harness: decisions, rules, and sprint context in a queryable DuckDB file, rules enforced on every model request, with a full management UI in the Web Client
Research lab for DeepSeek Harness: AutoSci-style wiki (experiments/literature/decisions), ASI-Bench-grade eval rigor ledger, arXiv digest, adversarial paper review, and academic writing checks.
Independent second-model reviews for DeepSeek Harness assistant decisions
Health monitoring, pressure scoring, maintenance scheduling and action decisions for DeepSeek Harness. It never restarts anything itself.
Project handoff skill: a dated handoff.md per project — current state, decisions with reasons, pitfalls as symptom/cause/fix, actionable next steps — so any fresh session resumes without archaeology. Bilingual EN/CN.
Triage a scattered backlog into a bounded queue of decisions: every item gets a disposition (now / next / scheduled / waiting / reference / drop) with a stated reason, dropped items leave the list instead of haunting it, and anything that changes a commitment goes out as one approval batch.
Lightweight decision records for any project: capture decisions that are expensive to reverse or re-litigate — decision, alternatives, the why, reversibility — append-only, supersede instead of rewrite, kept where the project lives so the next session inherits the reasoning, not just the conclusion.