/output

description: "Generate output artifacts from active wiki content — summaries, reports, study guides, slide outlines, timelines, glossaries, comparisons. Outputs are filed back into the wiki."
argument-hint: "<type> [--topic <topic>] [--sources <paths>] [--with <wiki>...] [--include-archived] [--wiki <name>] [--local]"
allowed-tools: Read, Write, Edit, Glob, Grep, Bash(ls:*), Bash(date:*), Bash(python3:*)

Your task

**Resolve the wiki.** Do NOT search the filesystem or read reference files — follow these steps: 1. Read `$HOME/.config/llm-wiki/config.json`. If it has `hub_path`, expand leading `~` only (not tildes in `com~apple~CloudDocs`) and prefer that path; use `resolved_path` only as a fallback cache when the expanded `hub_path` is unavailable and `resolved_path` is initialized. If config has only `resolved_path`, use it. If the configured path can be statted but reading `wikis.json` or listing `topics/` fails with `Operation not permitted`, stop and ask the user to grant Full Disk Access/iCloud Drive access to the launcher; do not fall back to `~/wiki` or `resolved_path`. Do not write machine-specific `resolved_path` into shared configs. 2. If no config → read `$HOME/wiki/_index.md`. If it exists → HUB = `$HOME/wiki`. If nothing found, ask the user where to create the wiki. 3. **Wiki location** (first match): `--local` → `.wiki/` in CWD; `--wiki <name>` → `HUB/wikis.json` lookup with portable path resolution (`<HUB>`, `~`, absolute, or HUB-relative); if the registry path is stale, fall back to `HUB/topics/<name>`; CWD has `.wiki/` → use it; else → HUB. 4. Read `<wiki>/_index.md` to verify. If missing → stop with "No wiki found (or no articles). Run `/wiki init` and `/wiki:compile` first."

Generate an output artifact from wiki content based on $ARGUMENTS.

Inventory and collect awareness: if the requested output is really a durable queue, candidate list, watch list, source backlog, or next-action table, recommend `/wiki:inventory` instead of creating another loose output. If the request needs web discovery plus a catalog of artifacts, examples, memes, media, tools, entities, or source candidates, recommend `/wiki:collect` instead. If an output report discovers durable follow-ups, include a short "Inventory candidates" section with a sample shape and ask before creating records. Do not bury ongoing tracking state in a generated report.

Parse $ARGUMENTS

• **type** (required): One of: `summary`, `report`, `study-guide`, `slides`, `timeline`, `glossary`, `comparison`. Use `/wiki:collect` for `type: collection` catalogs because those require discovery, dedupe, media policy, and inventory fit checks.
• **--topic <topic>**: Focus on a specific topic or concept (matches against tags and titles)
• **--sources <paths>**: Comma-separated list of specific wiki article paths to use
• **--with <wiki>**: Load a supplementary wiki as additional context. The primary wiki (`--wiki`) provides the **subject** (what to write about); `--with` wikis provide **craft/skill** knowledge (how to write it). Multiple `--with` flags allowed. Example: `--wiki quantum-computing --with article-writing` uses quantum-computing for domain content and article-writing for structure, hooks, and writing techniques.
• **--retardmax**: Ship it NOW. Don't agonize over structure or completeness. Generate a rough-but-useful output fast, covering everything the wiki has. Better to have something imperfect now than something perfect never.
• **--include-archived**: Explicitly allow archived primary or supplementary

wiki context. Label archived-derived material in the generated output.

Archived wikis are excluded by default, including from `--retardmax`. If the primary target or a `--with` wiki is archived, stop unless `--include-archived` is present. Never silently mix archived citations into an active output.

Output Types

**summary**: A condensed overview of a topic or the entire wiki. 1-2 pages max. Key points, main takeaways.

**report**: A detailed analytical report with sections, evidence from articles, and conclusions. 3-5 pages.

**study-guide**: Key concepts with definitions, questions and answers, concept relationships. Designed for learning.

**slides**: A markdown slide deck using `---` as slide separators. Each slide has a title, 3-5 bullet points max. Suitable for Marp or similar renderers.

**timeline**: Chronological view of developments, events, or milestones in a topic area. Formatted as a dated list.

**glossary**: Definitions of all key terms found in the wiki (or a topic subset). Alphabetically sorted.

**comparison**: Structured comparison of 2+ concepts/technologies/approaches. Uses tables for feature-by-feature comparison.

Retardmax Mode (`--retardmax`)

When `--retardmax` is set:

• Read ALL articles in the wiki, not just topic-matched ones. Don't filter, don't be selective.
• Generate the output immediately. Don't plan the structure first — just start writing.
• Include everything that might be relevant. Too much is better than too little.
• Don't worry about perfect formatting or transitions. Get the content down.
• If it's a report, make it comprehensive and raw. Polish comes later.
• If it's slides, more slides is fine. Trim later.
• File it and move on. The user can iterate with a non-retardmax pass.

Process

1. **Gather sources**:

• If `--retardmax`: read ALL wiki articles across all categories
• If `--sources` provided: read those specific articles
• If `--topic` provided: search wiki indexes for matching articles by tag/title, read them
• If neither: read `wiki/_index.md` for an overview, read articles from each category

2. **Load supplementary wikis** (if `--with` specified):

• For each `--with <name>`, look up in `HUB/wikis.json`
• Read the supplementary wiki's `_index.md` and relevant articles
• These provide **craft/skill context** — techniques, frameworks, writing patterns, best practices
• The primary wiki provides the **subject matt