Plugins

Drives a Chromium browser you are already running — Chrome, Edge, Brave, Vivaldi, Opera, Chromium, Arc or Thorium — over one long-lived CDP connection, so the remote-debugging approval is asked once instead of per command. Commands for tabs, navigation, JS evaluation, DOM snapshots, clicks, typing, screenshots and CSS/DOM/pixel assertions go over a loopback HTTP route that the agent calls with the shell tool it already has, so no new tools are registered, and the plugin registers its own usage skill with no runtime dependencies.

Compares existing DSH sessions or runs controlled A/B trials between agent presets, showing aligned timelines, paired metric deltas, explicit success checks, and exportable reports.

Découverte, évaluation, hiérarchisation adaptatives des modèles et routage des sous-agents pour DeepSeek Harness.

Observabilité Langfuse pour DeepSeek Harness (DSH) : arbres de traces par tour agrégés par identifiant de session, avec versionnage des prompts et évaluation.

Rejoue des tours réels de DeepSeek Harness contre des candidats Standard, Minimal, Anchored ou plugin, avec des preuves figées de la surface de requête

Vérifiez quels plugins DeepSeek Harness sont réellement chargés et consultez les tableaux d'évaluation publics sur trapstreet.run.

Pont CDP vers un Chrome/Edge déjà en cours d'exécution pour DSH : captures d'écran, assertions au pixel, vérifications DOM/CSS et évaluation de JS de page via HTTP + WebSocket, sans processus enfant, sans dépendance npm et sans approbation à chaque usage.

Registers one `valuation_service` tool that reads a workpaper tree of 01/02/04 JSON and returns one unified schema: six ratio groups, forecast ratios, DCF input slots with explicit missing markers, comparable PS/PE and IRR. Deterministic and local: missing WACC, growth, capex, D&A or working-capital inputs stay blank instead of being guessed, and assumption differences are flagged for review.

Checks a bid evaluation scoring matrix for internal arithmetic consistency: that each factor's maximum, the scores recorded and the total agree with each other. Reports the differences it finds; it does not judge whether the evaluation itself was correct or fair.

Cross-session memory for your dsh agent: remembered, recalled, and re-verified instead of assumed.

DSH 试金石：给「自改造」补上评测这一环——用一套金标准用例，把改动前 vs 改动后跑一遍、打分、对比，好就留、不好就撤。

Retrospective agent-session evaluation: deterministic grade cards and cross-session regression diffs from persisted logs — no benchmark runs, no LLM judges.

Measure whether a dsh setup change actually helped: register repeatable cases, run them, and diff the results before and after you change rules, skills, or models.

DSH plugin for bounded agent evaluation: compare an original with candidate prompt or supported configuration changes against user-supplied checks, record evidence and costs, and leave adoption to the user.

DSH web plugin: a LiveBench tab in the Trajectory view (right of 对话/轨迹). Run LiveBench evaluations against every model configured in the DeepSeek Harness — pick provider/model, category, task, release and question range from dropdowns, watch progress, and

Paired experiments and promotion gates for DSH plugins.

Runs one coding prompt across multiple configured models in isolated Git worktrees, streams side-by-side tool traces and results, and commits the candidate the user adopts for later rounds.

Evidence-first multi-model coding comparison workbench for DeepSeek Harness

Audit an agent harness against the harness-evaluation criteria, with machine-enforced evidence validation.

Exporte les événements de tour, modèle, outil, approbation, nouvelle tentative et erreur de DSH (métadonnées uniquement) vers Tuning Engines pour des traces gouvernées, une évaluation des politiques, une analyse des coûts et une revue des Work Sessions, avec une file de nouvelles tentatives persistée sur disque.

Plateforme d'évaluation d'agents : benchmarks YAML, orchestration d'exécutions headless, métriques basées sur les traces, et rapports d'exécution.