Plugins

Audita las skills y plugins instalados: asigna una herramienta según la descripción de la tarea, detecta desfases de versión frente al registro npm y fallos de ejecución en logs del host, e informa de herramientas de plugin que nunca se usan. Registra 9 tools: can, find, status, recommend, check, scan, probe, run y suggest. Funciona con Python 3.8+ sin dependencias de terceros. Todos los IDs usan el prefijo skill_audit_ (p. ej., skill_audit_can y skill_audit_find).

Iter Memory recuerda mi estado, no los hechos. El nivel vuelve solo al punto base y una frase elegida solo se dice a veces cuando ese nivel lo permite. Si está en silencio, no gasta nada.

Pantalla inicial de inmersión abisal para DeepSeek Harness: medidor de profundidad con sonda sumergible, vídeo de presentación del personaje en bucle y fondo marino con coral, burbujas y partículas lentas. La profundidad alcanzada determina qué criaturas aparecen.

Muestra marcas de proveedor y lista sincronizada: replica catálogo/logos públicos en loopback y regenera los modelos de pi-ai desde su endpoint, mostrando diferencias antes de escribir.

Tema celeste adaptado a la ciudad local: sol en movimiento, colores de día y noche, nubes flotantes, vista previa por hora y cielo inmersivo en el workspace web DSH.

Muestra cambios de archivos por turno, diferencias por archivo y veredictos de seguridad deterministas para sesiones DSH. Incluye un retroceso primero con vista previa que se niega si el worktree cambió desde la vista previa.

Guarda de consistencia entre documentación y código: audita encabezados de versión, tablas de cambios, árboles de directorios, recuentos de módulos y pruebas y referencias cruzadas entre documentos en busca de desviaciones, con sugerencias de corrección de solo lectura ordenadas por severidad.

Mide reglas de diseño en la página renderizada en lugar de en el CSS — número de escalas tipográficas, contraste WCAG AA sobre el fondo realmente compuesto, deriva de la rejilla de espaciado, tamaño de la paleta, objetivos táctiles y los indicios de UI generada como los degradados violetas detectados por ángulo de tono; incluye una skill de sistema de diseño.

El gestor de plugins `dsh pm`: búsqueda multifuente (lista awesome + GitHub + npm), instalación/eliminación/actualización por perfil, y una auditoría de tipo doctor de manifiestos, parches de paquetes y desviaciones de versión.

Freezes a six-step repair chain into callable tools — freeze the facts, measure the real shape, minimal repro, single-variable A/B, a judge that must say OK, delivery gate — plus a zero-dependency boot self-check (path drift, missing files, version sync, crash-log freshness), per-item tunables and a rollback ledger that backs up every action.

Storage and dependency inspector for the DSH Web UI: scans disk use per workspace, session and file class, lists the largest folders and files, and deletes the build, dependency, archived-session and cache folders you select behind a path allow-list (cleanup candidates also include top-level Desktop folders); a read-only second tab compares package.json, .nvmrc, .node-version and npm lockfiles across workspaces and flags version and Node-major drift.

Capability drift detection: audit a plugin once, store a baseline of what it could do, then after every upgrade report only what is NEW (new install scripts, shell/eval access, outbound domains, credential paths). Deletions and version bumps are not reported, and no score is computed.

Measures how much a short clip changed and when — changed-area share, displacement, cumulative drift — returning one labelled key-frame image plus the number behind each event; it does not pick shots for a storyboard. Needs Python 3.10+ with numpy and Pillow; reading mp4 also needs ffmpeg and ffprobe on PATH.

Replaces the composer model picker with a model + thinking-effort picker and an Auto level; Auto scores each turn by how far the conversation has drifted from the main objective and by repeated failures (one extra small judge call per turn), picks the effort from that score, and labels every finished turn with the effort it sent.

Declares a UI layout contract, measures pages in headless Chrome or Godot scenes and declared layouts, and audits overlap, occlusion, overflow, clipped text, undersized targets, dead controls, spacing drift and z-order traps, with a breakpoint matrix and a sidebar panel. Page measurement needs Chrome, Edge or Chromium and the puppeteer-core package installed where the DSH profile can resolve it; Godot scenes and declared layouts need neither.

Quote an earlier turn of the current conversation into the composer from the @ menu — by turn, not by selected text — so a chat that drifted can be re-anchored on what was actually asked.

Audits and evolves agent context at four scopes: a single repository, a group of related repos, a knowledge hub, or the whole machine. Discovers projects, instruction files, skills, tools, knowledge, project state, evidence and credential references; reports drift, duplicate facts, broken pointers, paths left unverifiable in prose, budget overruns, unindexed knowledge, stale state and unregistered credentials; scores each scope on the dimensions that belong to it; and turns findings into reviewable migration plans that back up every file before moving it. The format contract behind the metrics is written down as a versioned spec (CTX rules, markdownlint-style IDs) that follows open conventions — AGENTS.md for instruction files, llms.txt-shaped link indexes. Also scaffolds a deliberately small hub for durable knowledge and project state. Ships a stdlib-only Python CLI plus two skills, with credentials referenced by path and never printed. Requires Python 3.9+ (stdlib only) and DSH 0.2.1-rc.2 or newer, below 0.3.0.

Ambient firefly forest behind the DSH Web UI: a full-window canvas with a dark shrub layer, breathing fireflies on grass blades, and an owl or a moth that lights up when a drifting spark passes; on DSH Desktop a workbench panel adjusts brightness, firefly density, owl frequency, season and UI transparency.

A merge of the dsh-effort-dial card structure and the reasoning-slider gradient/particle effects: a Codex-style reasoning-effort card in the composer with one control for model and effort, per-level accent colors, a quota line on the top level, and a blue-purple gradient with drifting particles. It takes over the composer model slot, so it cannot be enabled alongside dsh-effort-dial or reasoning-slider.

Give your agent an external emotion state machine: several axes (0-100) that drift over time and move when you speak to it. Each turn the plugin injects which band you are in plus which tags you may use. The numbers belong to the plugin: the model reads them from the injected prompt and can change them only through the plugin's own emotion tools, never by asserting a mood.

Injects a Chinese-thinking reminder around the system prompt and the per-step runtime snapshot, so the agent reasons in Simplified Chinese instead of drifting back to English as the transcript grows.

Barra de estado para la barra lateral de DSH. Mientras se ejecuta una sesión, la fila se vuelve una tarjeta verde con una barra de progreso basada en el tiempo estimado del turno (pista punteada en movimiento) y un porcentaje en vivo en el extremo delantero. Al superar la estimación, muestra el tiempo transcurrido mm:ss. Al terminar, la fila se vuelve naranja y permanece iluminada hasta que abres esa conversación. Solo funciona en el cliente, sin servicios adicionales, y usa unas pocas claves de localStorage.

Herramientas ai-code-quality (cq) para DeepSeek Harness: pipeline determinista doctor/scan/triage/baseline/diff/gate/waiver/spec-drift que agrega ruff, mypy, radon, vulture, pytest-cov, import-linter y DeepSec, con análisis de solo lectura.