Chat Interface & Slash Commands
Relevant source files
The following files were used as context for generating this wiki page:
- bundled-skills/wiki-onboarding/SKILL.md
- bundled-skills/wiki-page-writer/SKILL.md
- bundled-skills/writing-skills/references/standards/README.md
- bundled-skills/codebase-audit-pre-push/SKILL.md
- bundled-skills/wiki-vitepress/SKILL.md
- bundled-skills/wiki-qa/SKILL.md
- bundled-skills/documentation-templates/SKILL.md
The Chat Interface and Slash Command system serves as the primary interaction layer between users and the OpenCode Skills Collection. It provides a structured environment where users can input natural language queries or specific execution directives (Slash Commands) to invoke specialized AI skills. This interface is designed to handle multi-turn conversations while maintaining a high signal-to-noise ratio through precise command triggers.
Within the project, slash commands act as specific entry points for complex workflows, such as /deep-wiki for site generation or /deep-wiki:onboard for documentation bootstrapping. This architectural choice ensures that AI agents can transition from general assistance to specific, specialized modes of operation with minimal overhead.
Sources: bundled-skills/wiki-onboarding/SKILL.md:12, bundled-skills/wiki-vitepress/SKILL.md:13, bundled-skills/wiki-qa/SKILL.md:12
Command Architecture & Trigger Logic
The system utilizes a metadata-driven trigger mechanism to identify when a specific skill should be activated within the chat interface. Each skill defines its own triggers and description within its Markdown frontmatter, allowing the chat interface to route user input to the appropriate logic handler.
Skill Activation Flow
The chat interface monitors input for specific keywords, error patterns, or explicit slash command syntax. Once a match is detected, the system loads the corresponding skill definition into the active context.
The diagram shows the transition from raw user input to specialized skill execution.
Sources: bundled-skills/writing-skills/references/standards/README.md:37-45, bundled-skills/wiki-qa/SKILL.md:18-22
Slash Command Definitions
Slash commands provide a shortcut for users to invoke complex, multi-step procedures without requiring descriptive natural language. These are defined within the description and metadata fields of individual skill files.
| Command | Skill Target | Primary Function |
|---|---|---|
/deep-wiki | wiki-vitepress | Builds a browsable HTML site from Markdown files. |
/deep-wiki:onboard | wiki-onboarding | Generates principal-level and contributor guides. |
/audit | codebase-audit-pre-push | Performs a production-readiness check on source files. |
/qa | wiki-qa | Answers questions grounded in source code evidence. |
Sources: bundled-skills/wiki-onboarding/SKILL.md:13, bundled-skills/wiki-vitepress/SKILL.md:14, bundled-skills/codebase-audit-pre-push/SKILL.md:12, bundled-skills/wiki-qa/SKILL.md:14
Execution & Context Handling
When a command is triggered via the chat interface, the system manages the "Documentation Budget" and context token limits. For high-complexity commands like wiki generation, the interface must handle asynchronous operations and data flow across multiple files.
Interaction Sequence
The following sequence diagram illustrates the communication between the User, the Chat Interface, and the Skill execution environment.
The sequence demonstrates the standard lifecycle of a slash command execution.
Sources: bundled-skills/wiki-page-writer/SKILL.md:33-36, bundled-skills/wiki-vitepress/SKILL.md:20-30
Content Standards & Verification
To maintain technical accuracy, the Chat Interface enforces specific response standards for all commands. Every response generated via a slash command must be grounded in evidence-based depth, ensuring that the AI does not "hallucinate" implementation details that do not exist in the source files.
Verification Checklist for Command Outputs
- Evidence-Based Depth: Claims must include file paths and line numbers.
- Traceability: Actual code paths must be followed rather than inferred from filenames.
- Formatting: Responses must use proper Markdown, including tables for API references and Mermaid diagrams for flows.
- Context Awareness: The interface must distinguish between verified facts and inferences.
Sources: bundled-skills/wiki-page-writer/SKILL.md:14-23, bundled-skills/documentation-templates/SKILL.md:46-52
Summary
The Chat Interface and Slash Command system provides the necessary structure for interacting with the opencode-skills-collection. By combining flexible natural language triggers with rigid slash command definitions, the system allows for both exploratory querying and deterministic workflow execution. This architecture ensures that specialized tasks—ranging from codebase auditing to automated wiki generation—are accessible through a unified, evidence-grounded communication channel.