Plugin

QA di fedeltà al design per modelli solo testuali: uno strumento `deepseek_vision` prende in prestito un occhio da qualsiasi percorso vision compatibile con OpenAI, così il modello può giudicare se un'implementazione corrisponde al mock — fornito con il benchmark alla base di quel giudizio (quattro fixture, 23 difetti iniettati, trascrizioni grezze) e la disciplina di interrogazione da cui dipende.

Configura il contextWindow per i modelli registrati in dsh, così le sessioni lunghe non vengono più erroneamente giudicate come overflow del contesto.

OMK — Osserva. Misura. Conosci. Modifiche della knowledge supportate da evidenze per applicazioni AI.

Mappa l'audio di sistema analizzato localmente su effetti di conversazione reattivi alla musica, giudizi di gioco ritmico, feedback di punteggio e combo e barre dello spettro RGB opzionali in DSH Web.

Harness di memoria dell’agente a tre livelli: CLAUDE.md L1 sempre attivo, indice L2, scheletro dettagliato L3 con modelli, un gate che blocca le scritture nei file protetti senza un token di approvazione monouso, registrazione degli eventi senza giudizi, distillazione settimanale e una skill di ciclo di esecuzione TDD.

Esegue 3 o 5 candidati agenti di coding in Git worktree detached, valida le loro patch con comandi di progetto, classifica solo i candidati che passano e richiede un'approvazione separata prima di applicare il vincitore.

Plugin modalità di apprendimento Feynman: ciclo insegna → ripeti l'insegnamento → valuta → rispiega, reso come schede lezione HTML ricche (diagrammi mermaid + evidenziazione del codice shiki)

Resumes an agent turn that ended after narrating its next action without calling a tool: the guard replays the turn log on agent/turn-stopping, asks a judge model one true/false question, and steers the same turn to run one more step.

Un sottoagente di approvazione indipendente giudica ogni escalation di sandbox, con un modello configurabile e un log di audit.

Automatic approval and review for tool calls that need approval: keyword buckets match red lines without context, then a judge model classifies what was requested and sends it to allow, reject or a human according to the matching row and risk level; a judgment that never ran always goes to a human, and rejections tell the model why.

Peak / off-peak rate badges for the model selector (every row) and the composer tool row, judged per provider by its own time zone and schedule; the live catalog is served from the host over a fenced route.

Commitment write-gate: operator-authored rules enforced before a tool call runs — a deterministic guard tier plus an LLM-judge tier, fail-closed by default, every block written to a contradictions log.

Self-learning habit engine for the DeepSeek Harness — detects user-correction signals, judges habits with a low-cost model on threshold, settles candidates behind a two-level human gate

Codex-style approval autopilot for DeepSeek Harness: ordered glob rules (allow/ask/deny) plus an AI risk judge (low/medium/high) mapped through a risk tolerance, as an approval answerer.

Autonomous permission classifier for the auto preset: tool-scoped allow/deny rules, an LLM semantic judge, and git checkpointing for unattended sessions.

Destructive-command interception gate for dsh: parses shell semantics, judges risk against 41 built-in rules, and holds irreversible rm -rf, git reset --hard, and git push --force style commands at a confirmation gate.

Router di modelli a due livelli per DeepSeek Harness — instradamento con LLM-Judge, catene di fallback multi-modello, failover con backoff esponenziale e orchestrazione a livello di task (adattamento DSH di pi-shift-router)

Cross-project experience recall for DeepSeek Harness: when the running work matches something your memory store already learned elsewhere, that lesson is injected before the next step. Runs fully locally, judged by a local small model.

DeepSeek Harness plugin bundle: grill_tree, a fixed-policy design-space search tool. Expands one decision per node, fans out independent proposers, has a separate judge prune the branches, beam-limits each depth, and ends with complete designs or an infea

DeepSeek Harness plugin for CVE impact retesting: fingerprint a target, judge it against a CVE's affected range, gather passive evidence, and report a verdict with confidence and limitations.

Resume screening plugin for HR in DeepSeek Harness: ingest up to 100k Word/PDF/Excel resumes into one searchable library with skills / coarse categories / ingest time, auto-ingest & de-dupe new resumes, then screen by plain-language criteria — rule pre-filter in the database plus LLM fine-judge on the shortlisted few, CSV export.

Shared toolkit for authoring DeepSeek Harness plugins, published as @perrylink/dsh-plugin-kit: a pluggable provider registry seam, fail-closed approval and session-event gates, shared sanitize/pricing/judge modules and a new-plugin skeleton.

Multi-role parallel argumentation for DeepSeek Harness: runs the real Codex and Claude CLIs in parallel, has the DSH model (Judge) write a third-party verdict back into the conversation, plus Obsidian-style single-agent direct chat and per-role model config.

DeepSeek Harness 插件：生成前注入「人味」风格约束 + AI 味检测报告（纯代码、零 token、结果可复现）。dsh plugin: human-flavor style injection + AI-flavor detection report, judged by code.