Plugins

Hire Google's Antigravity CLI (agy) as a fast Gemini staffer for Claude Code, OpenAI Codex, and DeepSeek Harness.

Design-fidelity QA for text-only models: a `deepseek_vision` tool borrows an eye from any OpenAI-compatible vision route, so the model can judge whether an implementation matches its mock — shipped with the benchmark behind that judgement (four fixtures, 23 injected defects, raw transcripts) and the questioning discipline it depends on.

Model selector for the DSH composer that keeps the native button and menu and takes over only when the small dropdown would open: search across providers, probe any model with a real streaming request to measure TTFT and TPS, and browse a leaderboard of up to 100 rows built from OpenRouter's catalogue, where clicking a row searches local channels of that name.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models. Within a turn it supports skeletonized information compression with a low original-text retrieval rate; at the tail of a turn it supports a turn-tail intent summary implemented purely by content substitution; context content stays safe, and the whole process runs inside the official JSONL compression.

OMK — Observe. Measure. Know. Evidence-backed knowledge changes for AI applications.

Talk to Excel in DeepSeek Harness: create, edit, repair, and verify spreadsheets by conversation, with automatic formula health checks after every edit.

Per-plugin usage budgets and cost caps on the llm/stream waterfall: per-model usage accounting, session and monthly budgets with warn/block transitions, latency windows, and a carbon-footprint estimate.

Zero-LLM auto-capture and a symbolic \[ENGRAM\] index with progressive disclosure, an ESR-lite task/evidence protocol (esr_task / esr_node / esr_close / esr_link), and per-workspace usage telemetry with an offline recall benchmark (npm run eval).

DSH tools for reproducing and validating the ml-quant-trading protocol v1 CPU benchmark, then drafting an issue-ready report. Requires a local clone of initial-d/ml-quant-trading (set as the workspace or repoPath) plus Python and PyTorch; this plugin does not fetch the repository or install its dependencies.

Talk to Excel in DeepSeek Harness: create, edit, repair, and verify spreadsheets by conversation, with automatic formula health checks after every edit.

Benchmark the current default model API and pick the fastest, most stable resolved IP node, optionally applying it.

Displays CodexRadar model capability benchmarks in a Settings page and shows the selected session model's DeepSWE score beside the composer.

Adds persistent execution contracts, recursive work graphs, critical clarification, and evidence gates for long or underspecified Harness tasks.

Sidebar leaderboard for DeepSeek Harness (DSH): LMArena blind-test ratings, Artificial Analysis intelligence index, LiveBench 2.0 capability, OpenRouter weekly/monthly usage and market share, price, speed and context — plus a local token-usage dashboard built from your own session logs (tokens, estimated cost, per-model ranking), model detail drawers and CSV export. Daily snapshot, instant open, no API key; locally configured models are highlighted.

Browser provider for DeepSeek Harness: hand it a web task — open a page, click, type, fill a form, read what is on it — and it drives a real browser step by step, seeing an indexed table of what can be done on the current page instead of raw HTML or screenshots. Each step is one bundled-Jev decision rather than a full model turn (1.1 s median per step versus 2.8–3.6 s for chat-model turns in the repo benchmark), and the main conversation receives only the conclusion, so its context does not grow with the page. Targets come from the page it just read and a changed page is re-read rather than clicked blind, and it runs in the browser that Session owns, leaving your everyday Chrome untouched unless you switch to attach mode.

Security-audit facts AI coding agents cannot compute, plus a verdict gate: euthyna history attributes deleted lines to commits and flags those from security fixes (--origins finds the first introducer), euthyna coverage reports which changed symbols no test ever invoked, and euthyna gate checks each finding against six gates, downgrading any without evidence to an observation.

Merges other sessions' context into the current turn: evidence-weighted conflict resolution, retraction of claims whose premises were retracted, trust labelling that resists cross-session prompt injection, and budget-bounded packing. Ships a benchmark.

Verification gate for DSH agents: every acceptance criterion must be backed by server-stamped real tool evidence before the completion gate lets a goal through (advisory audit, enforce gate, durable permits).

Inspect API latency, streaming output and model-response fingerprints from a native DSH panel, with classic-task previews and public benchmark rankings.

dsh bundle for TSM / DLR: MCP semantic gateway to tsm-core-dlr, the scenario-independent paradigm skill, the TSM status overlay, and the agent rules (AGENTS.md). Assembly face = preset-dlr.

Runs a fixed set of agent benchmark cases in isolated sandboxes, scores each case with deterministic assertions, and diffs two runs to show which cases regressed after a model, prompt or DSH upgrade.

Evaluation guard for DSH: workspace-only filesystem, no network/subagents, wall-clock limits, and task installation for meow-ai-arena and benchmark tasks.

Mounts the tsm-core-dlr semantic service (LanceDB vectors + graph) as MCP tools and ships the scenario-independent DLR/TSM modeling skill, a service-status overlay and the agent rules; the service is a hard dependency and must be running before the bundle activates.

0.8.6 fixes a startup crash: the host runtime package is an optional peer, never an installed dependency. TypeSafe Jev (System One decision model) wired end to end into DeepSeek Harness (0.8.6). Seven tools: jev_decision (1-25 parallel atomic verdicts in one call; noul/choice/score; typed answers with calibrated confidence; measured median 266-484 ms; verdicts only, never prose), jev_choose (rank 2-10 candidate approaches - fit 0-3 and risk - into an explicit composite with a recommended pick), jev_batch (NEW in 0.8.6: one set of 1-5 judgments over 1-20 texts, one real call per text at bounded concurrency, returning a paste-ready markdown table sortable by a judged dimension; built-in sentiment/priority/intent/pii/spam presets; a failed row is reported as failed with the raw error, never guessed), jev_usage (locally measured usage: calls, tokens, estimated cost, latency median and p95, per-tool mix, rolling windows, projected burn rate, optional local hard budgets, and a declared-balance remainder), jev_overview (session ledger), jev_guard_status (guard counters plus a registrations block that surfaces per-surface registration failures) and jev_verify (frozen 27-question live benchmark at concurrency 6: 96.3%, 26/27). WHAT IS NEW IN 0.8.6 - the ledger survives a restart, the remainder is derived instead of invented, and the plugin can now do batch paperwork. (1) The local ledger persists to $DSH_HOME/jev-usage.json BY DEFAULT, so the cumulative total no longer resets when the host restarts. Two real bugs behind the old silent zero were found and fixed: the ledger path only honoured DSH_HOME (absent inside the host process, so persistence was silently disabled and the file name was null), and save() wrote the still-empty in-memory ring over the history file because it ran before the first load while the debounce window swallowed the follow-up write. (2) Remaining is derived, never invented: TypeSafe exposes no balance endpoint - /v1/usage, /v1/quota, /v1/account, /v1/account/balance, /v1/balance, /v1/credits, /v1/billing, /v1/me, /v1/limits and a dozen more all return 404, and only /v1/models answers, with no balance headers - so the panel shows the remainder as your declared balance (quota.declaredBalanceUsd) minus locally measured spend since quota.balanceSince, and says plainly that it cannot be fetched instead of printing a number. (3) The composer pill re-reads every 20 s instead of every 60 s, so a call you just made shows up while you are still looking at it. (4) The collapsed pill leads with the remainder once you declare a balance, and the cumulative row is renamed when the ledger is on disk. Every tool has an in-conversation structured view; the Settings > Plugins card covers credentials, tool switches, the auto-guard, the dashboard, the prompt section, usage/quota and auto-triage. Honest by design: no API key means an explicit error, never fabricated results; every registration is quarantined so a failure degrades one surface. Tests 38/38 (npm test, ~2.5 s). Client half applies on refresh; server half needs a host restart.