haowjy

Load for large human-facing outputs: reports, docs, multi-section explanations, artifacts. Choose the right form for each beat (prose, diagram, table, mockup) and put the answer first, depth behind it. If the medium allows, use progressive disclosure.

Use when deciding where knowledge goes or reading/writing durable docs: AGENTS.md, .context/, KB, docs/, and work directories.

Use when exploring or changing the codebase: read AGENTS.md first, use .context/CONTEXT.md for detail, keep intent docs succinct.

Load before writing or revising human-facing text. Choose words deliberately, ground the piece in the reader's context, and remove default LLM phrasing before the final draft.

Use when validating markdown links or Mermaid diagrams.

User-invoked pause before reporting to check intent vs literal completion, surface adjacent wins, and route knowledge capture.

Load when building a static HTML artifact — single page or multi-page site — to make structured information explorable through layout and navigation.

Use when orienting in an unfamiliar area — fans out exploration across structure, knowledge, and history, then synthesizes one orientation map.

Use when grilling should be grounded primarily in project artifacts. Loads `/grill`, checks project vocabulary and prior decisions first, and spawns bounded explorers when the answer should come from docs.

Load for large human-facing outputs: reports, docs, multi-section explanations, artifacts. Choose the right form for each beat (prose, diagram, table, mockup) and put the answer first, depth behind it. If the medium allows, use progressive disclosure.

Use when deciding where knowledge goes or reading/writing durable docs: AGENTS.md, .context/, KB, docs/, and work directories.

Anchor work in a work item: artifacts on disk, not in context.

Investigate and engage concurrently: don't block on evidence before surfacing your read.

Use when exploring or changing the codebase: read AGENTS.md first, use .context/CONTEXT.md for detail, keep intent docs succinct.

Use when recovering decisions or context from conversation history: session log navigation, transcript delegation.

Use when establishing shared vocabulary: resolve terminology early, one name per concept, vocab files as record.

Multi-model supervisor that discovers skills, picks models, and composes runs. Use when executing multi-step plans across multiple models.

Agent execution engine that composes prompts, routes models, and writes run artifacts. Use when launching subagent runs.

Multi-model supervisor that discovers skills, picks models, and composes runs. Use when executing multi-step plans across multiple models.

Agent execution engine that composes prompts, routes models, and writes run artifacts. Use when launching subagent runs.

Verifies implementation aligns with stated requirements and acceptance criteria. Use before implementation (to clarify scope) and before final sign-off (to detect gaps).

Breaks the next task from a plan into an implementable task file. Use when decomposing a multi-step plan into ordered work units.

Explores codebases and evaluates approaches before planning. Use when investigating a problem space, comparing alternatives, or gathering context for a plan.

Reviews code against project rules and curated reference files. Use when auditing files for violations, reviewing changes, or generating cleanup tasks.

Load when writing or reviewing agent .md files. Covers agent artifact shape, frontmatter, and Meridian-specific agent conventions.

Load when writing or reviewing prompts, skills, or agents. Gives the core heuristics for writing prompts that steer LLM behavior reliably.

Load when writing or reviewing skills. Covers SKILL.md shape, loading mechanics, and how to structure bundled resources.

Load when an agent, skill, or prompt needs strict review. Focuses on failure modes, severity, and clear feedback.