harnessprotocol

Use when user invokes /capture to capture information from the current conversation into a staging file for later reflection and knowledge graph processing. Accepts no arguments (auto-extract), specific facts as text, or filter keywords like "decisions" or "technical". Do NOT use when the user wants to write directly to knowledge files — use direct file editing instead.

Use when user invokes /lineage with a column name (optionally qualified with table/schema). Also triggers on "trace this column", "where does X come from", "what reads from Y table". Traces column-level data lineage through SQL, Kafka, Spark, JDBC, and ORM codebases. Produces a structured lineage path with confidence ratings and an SVG or ASCII diagram. Do NOT use for general code explanation — us

Use when user invokes /review to review code changes. Also triggers on "review my code", "check this PR", "look at my changes". Accepts a PR number (e.g. /review 123), a path filter (e.g. /review src/auth/), or no argument (reviews current branch vs base). Produces structured per-file review with severity labels, cross-file analysis, and an overall verdict. Do NOT use for explaining what code does

Use when user invokes /orient with a topic keyword, entity type, project name, time qualifier, or combination. Also triggers on "what do we know about X", "remind me about X", "where did we leave off on X". Provides targeted context loading — searches the MCP Memory Server graph, knowledge files, journal entries, and research index to produce a focused orientation briefing.

Use when user invokes /memory with a subcommand (search, trace, add, episode, status). Also triggers on "search my memory for X", "add to memory", "what do I know about X" (when membrain is available), "memory status", "memory trace". Searches and manages the membrain knowledge graph via MCP tools or HTTP API.

Use when user invokes /harness-export or wants to save their current harness-kit plugin setup to a shareable harness.yaml file. Detects installed skills, collects source info, and writes the config in Harness Protocol v1 format. Do NOT use for importing or restoring a harness — use /harness-import instead.

Use when user invokes /harness-import or wants to install plugins from a shared harness.yaml config file. Reads the config, presents each plugin interactively, and generates the Claude Code install commands for the ones the user selects. Handles both Harness Protocol v1 format (version: "1") and the legacy format (version: 1).

Use when user asks to build a web component, page, or application, or when the task involves frontend design, HTML/CSS generation, or UI layout. Applies specific rules for typography, OKLCH color, layout, motion, interaction, and UX writing to produce distinctive, production-grade interfaces that avoid generic AI aesthetics.

Harness Kit documentation — installation, plugin catalog, creating plugins, cross-harness setup, architecture, and FAQ. Use when working with or configuring harness-kit plugins, understanding the plugin/skill system, installing slash commands, setting up AI coding tool configuration, answering questions about the plugin marketplace, writing SKILL.md files, using harness.yaml, or integrating with C

Use when user invokes /refresh-research to re-fetch refreshable research sources, detect meaningful changes, and update synthesis files. Supports specific files, categories, staleness thresholds, backfill of source_type, and audit-only dry runs.

Use when user invokes /harness-compile or wants to compile a harness.yaml into native config files for Claude Code, Cursor, and GitHub Copilot. Generates CLAUDE.md, AGENT.md, .mcp.json, .cursor/rules/, .vscode/mcp.json, and related files from a single harness.yaml source. Supports --target, --dry-run, and --clean flags.

Use when wrapping up a development task and getting a PR ready — runs local tests, creates a PR (if one doesn't exist) with a structured description template, conducts a code review via the review skill, and checks CI status with quick fixes. Trigger when the user says they're done with a feature, want to open a PR, wrap up, finalize their work, or push their branch for review. Also invoke proacti