Plugins

Outils d'humain dans la boucle (`human_task` / `human_task_ready_check`) : mettent l'agent en pause jusqu'à ce que l'utilisateur effectue une action dans le monde réel ou renvoie une observation (opération GUI, test de jeu, vérification visuelle, contrôles matériels), conditionnés par le consentement de la session et une vérification de présence AFK, renvoyant un résultat JSON structuré via une boîte de dialogue web de tâche.

Chercheur de plugins DSH conversationnel : index unifié de quatre sources, pré-filtre BM25 plus classement sémantique LLM, les candidats portent des grades de preuve et un audit de sécurité statique local, flux d'installation sûr avec vérification post-installation.

Workflow d'ingénierie full-stack natif Cordis et fondé sur les preuves : planification qui inspecte d'abord, preuves de vérification explicites et portes tenant compte des approbations pour les opérations sensibles.

Motor-design AI assistant plugin: requirement parsing → parameter identification → candidate matrix → Top10 verification. L0 is a millisecond-level broad filter (pure JS, zero solver dependency) that ranks candidates and emits the handoff payload for the L1/L2 precision tools; usable as a local agent preset and Cordis plugin. L1 (RMxprt) and L2 (Motor-CAD) are reserved in this version — the level gate fails fast above l0.

Patch-style file editing: one apply_patch tool accepting git/unified diff (default) and Codex apply_patch syntax (opt-in), applying every change all-or-nothing after a full read-only verification pass, with sandbox-aware delete and move and a dry-run mode.

Verifiable authorization, execution evidence, and offline delivery verification for DeepSeek Harness

Deterministic Reasoning Engine (DRE) for dsh: dre__* tools for knowledge verification (three-stage discrimination), deterministic cognition loops, constraint solving, mental models and synapse memory - strengthens information certainty.

Linux X11 computer use for DeepSeek Harness: 9 tools (screenshot, move/click/drag/scroll mouse, type, key combos) driven by a zero-dependency Python XTest helper — native-pixel coordinates, per-action verification, optional auto-observation screenshot.

AI Agent runtime authorization & evidence verification — tool-call GuardrailProvider, CCS 7-dimension verification standard, MCP/DSH security scanner, SSRF/command-injection/credential-exfil blocking with Ed25519 signed receipts.

LLM-as-a-Verifier for dsh with the Best-of-N conversation mode built in: rank N candidates with a fine-grained verifier (expected grade over the logprob distribution), on demand via the verify tool or automatically on every turn of a Best-of-N session. In

Runtime governor for DeepSeek Harness: zero-progress steering, generation-bound verification evidence, evidence-aware completion guarding, and task-profile tool restriction.

Model-agnostic visual-inspection pipeline for text-only agents: page-by-page HTML screenshots plus a ready-made vision-subagent briefing contract (gewu_prep), then source-code truth verification of every finding (gewu_locate); validated on mimo-v2.5 & qwen3.7-plus.

DeepSeek Harness community Bundle for the BeforeDone evidence gate

Verification-driven self-evolution loop: failure logs become verified AGENTS.md rules; in-session plugin (evolve_learn / evolve_apply / evolve_touch / evolve_recall), tool-verify gate, rule lifecycle, local recall.

Verification toolkit for DSH agents: evidence-based claim checking against workspace files with line citations, config validation (JSON/YAML), and read-only URL/npm/GitHub submission-readiness probes.

Headless-browser verification tools so the agent opens the page it just built, reads the rendered DOM and computed styles, checks the console, and screenshots the result; ships a frontend-verify skill.

Human-in-the-loop tools (`human_task` / `human_task_ready_check`): pause the agent until the user performs a real-world action or returns an observation (GUI operation, game testing, visual verification, hardware checks), gated by session consent and an AFK presence check, returning a structured JSON result via a web task dialog.