Explicit model input-modality selection for the DeepSeek Harness Web UI

DeepSeek Harness Web plugin: per-model input-modality selector on the Models settings page for third-party (pi-ai) providers

Mobile UI fixes for the DeepSeek Harness web profile: a restart row, input-modality switches, a narrow-screen drawer and settings tabs, a full-width model menu, a stuck-request network chip, a scrollable question region, a throttled reaction to streaming

Third-party provider reasoning-effort AND input-modality settings for DeepSeek Harness: thinking levels and image-input support declared per model, auto-adapted from a model knowledge base + wire-protocol inference, edited right inside the official Models

Resolve a URI relative to an optional base URI

Pi agent extension that parses image/audio/video inputs and forwards them to an OpenAI-compatible multimodal proxy (e.g. dots.ai).

Model-facing describe_image tool for the dsh web GUI: gives a text-only model image understanding by asking a vision-language model at an OpenAI-compatible endpoint to describe one image (local path, http(s) URL, or attachment reference). Hot-pluggable —

One-time password input component for React.

Per-model capability declarations (image input and reasoning efforts) for custom pi-ai providers, edited in place on the Models settings cards; writes the official llm-pi-ai settings namespace, no DSH source changes.

Get the first fulfilled promise that satisfies the provided testing function

Isomorphic, functional type-checking for Javascript

Framework-agnostic trait model and adaptation policy engine for Protean UI

Inquirer input text prompt

A family of specs for interoperable TypeScript

Convert a dash/dot/underscore/space separated string to camelCase or PascalCase: `foo-bar` → `fooBar`

Match human-quality input to potential matches by edit distance.

JavaScript parser, mangler/compressor and beautifier toolkit

Parse postgres array columns

Merge CSS rules with PostCSS.

Retry a promise-returning or async function