DSH Better Reasoning Effort

English | 中文

Reasoning-effort and input-modality editing for third-party models in DeepSeek Harness, edited right inside the official Models page card — plus a quick reasoning-effort slider inside the official composer model menu (adapted from HanaAyane's dsh-reasoning-effort, see Acknowledgements).

Why

The llm-pi-ai adapter natively supports per-model reasoningEfforts and input declarations, but the official Models page editor deliberately keeps both fields out of reach. As a result, third-party models get no thinking-level picker in the composer, only the official DeepSeek API can set reasoning effort, hand-declared models are treated as text-only, and configuring any of this meant hand-writing settings.yaml blocks. This plugin brings both configuration surfaces back into the UI: edit inside the official model editor card, plus one-click auto-adapt.

Features

• In-page editor — an "Options" block appears under each model row's disclosure on the official Models page (reasoning effort, input modalities, endpoint compatibility), committing with the card's own Save; changes stay pending until then, and Cancel discards them with the card's fields.
• Auto-adapt — one click fills recommended levels, wire spellings and modalities from a built-in knowledge base (65 entries across 15 vendors — see Supported models), wire-protocol inference, and a same-origin probe of the provider's raw /models listing, every suggestion labeled by confidence; reference capacities show as read-only hints you copy yourself.
• Auto-fill — models with a curated official knowledge-base match are filled at boot and when added mid-session (opt out via autofill: false / modalityAutofill: false); unknown models stay unchanged until the user explicitly accepts Auto-adapt.
• Three intents — all levels off = unset (back to inheritance); only off = disable reasoning; levels armed = write the declaration.
• Catalog-safe writes — routes using DSH's official modelOverrides keep their installed catalog; only the edited model override is written.
• Composer slider — the official model menu's body is replaced by an upstream-style effort slider (drag / keyboard, optimistic commit with rollback), with one model row opening the official model list; the official trigger stays untouched, and model switches keep your level through a per-session memory chain (issue #4's per-model default effort outranks it across sessions).
• Per-model default effort — a "Default effort" picker on each model row, stored in the settings document; every new session starts the model there.
• Request headers & user-agent — a provider-card section edits the official headers field (masked, path-merged, Save-gated), and the plugin performs the user-agent override at the fetch layer per origin, because the official adapter reserves that name; same-origin /models probes are covered, conflicts are reported rather than guessed.
• Defensive injection — everything keys off the official page's DOM; if an official upgrade changes the structure, injection simply pauses and the official page is unaffected.
• Bilingual copy (中文 / English).

Supported models

…