القائمة
Guidelines for ensuring feature parity and core principles when developing Dux User Interfaces.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Guidelines for creating, reviewing, and improving agent skills.
Guidelines for validation, staging, amending, and commit messages.
Guidelines for dependency injection patterns.
Guidelines for ensuring architectural changes are validated against core use cases.
Orchestrates efficient read and write interactions with the engram unified memory store via MCP tools. Provides deterministic workflows for decomposing observations into atomic memories, composing four-dimensional queries (context, similarity, relationship, time) for maximum retrieval accuracy, managing agent-managed focus context, and linking memories into associative webs at write time. Ensures memories are discoverable, precisely ranked, and structurally connected across sessions.
Guidelines for dual-use architecture (CLI and Go Library), dependency separation, and extensible registry patterns.
| name | ui |
| description | Guidelines for ensuring feature parity and core principles when developing Dux User Interfaces. |
When designing, modifying, or creating user interfaces for the Dux application (e.g. Web Apps, Terminals, IDE extensions), you must adhere to a core set of standardized pillars. These pillars guarantee feature parity between our clients and ensure that the complex background behavior of our agentic system remains observable.
Dux is built around iterative text and structured-data streams. Interfaces must eagerly consume and render stream chunks ([]llm.Part) as they arrive.
Advanced agents perform hidden logic. A Dux UI is responsible for visualizing this background execution natively as disjoint elements from the raw Assistant Response.
llm.ReasoningPart) and place them in a visually distinct bounding box/element near the top of the interaction. Do not merge them into the assistant's standard text message.llm.ToolRequestPart and llm.ToolResultPart). Render the names of the tools and their payload/responses so the user knows what background services were interrogated. Important: Because the LLM continues reasoning in a connected internal loop after a tool finishes, you must cleanly interrupt the visual stream so that the assistant's final answer appears natively after/below the visualized tool call, rather than merging back into the preamble above it.llm.TelemetryPart). Show token usage (Input, Output, Reasoning) and overall query duration sequentially.An unpredictable AI must be gated appropriately before mutating system state.
When building a new UI or assessing an existing one, check it against the definitions above. Regardless of visual style, all Dux UIs must expose the above functional phenomena. If functionality like Telemetry Tracking or explicit Background Tool Visualization is missing, treat it as a critical parity gap and address it.