Skip to content

README.md

Latest commit

 

History

History
316 lines (222 loc) · 10.5 KB

README.md

File metadata and controls

316 lines (222 loc) · 10.5 KB
Raw

Example Maps

These example maps are meant to be opened, searched, filtered, and deep-linked, not just skimmed.

Each one demonstrates a different style of thinking in mdmind:

  • product and roadmap planning
  • model comparison and benchmark review
  • meeting notes and action follow-up
  • prompt iteration and evaluation
  • decision logging
  • agent TODO maps and handoff loops
  • agent-to-human research handoff
  • studio operating maps
  • game worldbuilding and quest architecture
  • novel research, themes, and drafting

You can explore them in two ways:

mdmind examples/lantern-studio-map.md
mdm view examples/lantern-studio-map.md#lantern/execution/now

Quick Picks

Maps

demo.md

The smallest map in the repo. Good for learning the file shape and basic tree navigation.

Try:

mdmind examples/demo.md
mdm view examples/demo.md
mdm links examples/demo.md --plain

product-status.md

A compact product planning map with features, active work, priorities, and docs-related branches.

What it demonstrates:

  • #feature, #todo, and status-style metadata
  • product breakdown by capability area
  • ids on important feature branches

Good first jump:

mdm view examples/product-status.md#example/product/features/docs

Good CLI probes:

mdm find examples/product-status.md "#todo @status:active" --plain
mdm kv examples/product-status.md --keys status,owner,priority --plain
mdm links examples/product-status.md --plain

model-benchmark-comparison.md

A model comparison and leaderboard cross-reference built to be opened as a metadata table rather than read as a deep hierarchy.

What it demonstrates:

  • a real column-first workflow for comparing AI models across multiple public leaderboards
  • rows as models with columns like @provider, @context, @aa_rank, @aa_index, @llms_rank, @llms_score, @lma_text_rank, and @lma_code_rank
  • explicit @match and @coverage fields for model-name ambiguity across sources
  • detail lines as source and variant notes, with all comparable data kept on the model rows
  • a shallow map that keeps non-row context out of the table body

Good first jump:

mdmind examples/model-benchmark-comparison.md

Then press C, press c, and select columns such as @provider, @context, @aa_rank, @aa_index, @aa_price_usd_1m, @llms_rank, @llms_score, @lma_text_rank, @lma_code_rank, @match, and @coverage.

Good CLI probes:

mdm kv examples/model-benchmark-comparison.md --keys provider,context,aa_rank,aa_index,aa_price_usd_1m,llms_rank,llms_score,lma_text_rank,lma_code_rank,match,coverage --plain
mdm find examples/model-benchmark-comparison.md "#model @provider:openai" --plain
mdm find examples/model-benchmark-comparison.md "#model @coverage:4" --plain
mdm view examples/model-benchmark-comparison.md#model-benchmark

meeting-notes-action-map.md

A classic meeting-notes outline with agenda, discussion notes, decisions, action items, and parking-lot topics.

What it demonstrates:

  • a traditional outliner-style use case that still benefits from tags, metadata, ids, and cross-links
  • longer discussion notes attached with detail lines instead of bloating row labels
  • turning one meeting into a reusable action surface instead of a dead transcript

Good first jump:

mdm view examples/meeting-notes-action-map.md#meeting/actions

Good CLI probes:

mdm find examples/meeting-notes-action-map.md "@owner:maya" --plain
mdm find examples/meeting-notes-action-map.md "#decision" --plain
mdm kv examples/meeting-notes-action-map.md --keys owner,status,priority,date --plain

prompt-ops.md

A prompt-iteration workspace for variants, evaluation notes, and prompt ownership.

What it demonstrates:

  • prompt families with sibling variants
  • metadata for evaluation and ownership
  • a structure that works for prompt libraries without turning into a spreadsheet

Good first jump:

mdm view examples/prompt-ops.md#example/prompts/variants

Good CLI probes:

mdm find examples/prompt-ops.md "@owner:jason" --plain
mdm tags examples/prompt-ops.md --plain
mdm kv examples/prompt-ops.md --keys owner,status,score,failure --plain

decision-log.md

A small decision record map for architecture or product tradeoffs.

What it demonstrates:

  • dated or staged decisions
  • pros/cons and consequences as branches
  • stable ids for citing decisions later

Good CLI probes:

mdm tags examples/decision-log.md --plain
mdm links examples/decision-log.md --plain
mdm find examples/decision-log.md "tradeoff" --plain

agent-research-handoff.md

A research synthesis map shaped the way an agent might hand off work to a human operator.

What it demonstrates:

  • agent-friendly map structure without over-structuring every branch
  • themes, evidence, open questions, and actions in one file
  • ids, details, and a few meaningful relations
  • a practical bridge from agent output to human cleanup in mdmind

Good first jump:

mdm view examples/agent-research-handoff.md#agent-handoff/themes

Good CLI probes:

mdm find examples/agent-research-handoff.md "@source:interview" --plain
mdm find examples/agent-research-handoff.md "#todo @status:active" --plain
mdm relations examples/agent-research-handoff.md#agent-handoff/themes/ownership --plain

agent-todo-workflow.md

A local TODO and handoff map for a developer working with Codex or another coding agent.

What it demonstrates:

  • decomposable tasks with #todo, @status, @owner, @priority, and @area
  • short handoff instructions that an agent can follow safely
  • validation and inspection commands embedded near the workflow
  • [ ] and [x] rows that toggle with Space in the TUI

Good first jump:

mdm view examples/agent-todo-workflow.md#codex-todo/current

Good CLI probes:

mdm find examples/agent-todo-workflow.md "task:open" --plain
mdm find examples/agent-todo-workflow.md "#todo @status:active" --plain
mdm kv examples/agent-todo-workflow.md --keys status,owner,priority,area --plain
mdm find examples/agent-todo-workflow.md "task:blocked" --plain
mdm find examples/agent-todo-workflow.md "@status:blocked" --plain

lantern-studio-map.md

A large, lived-in studio operating map built around the fictional immersive nighttime experience Drift Signal. It mixes creative direction, route design, launch prep, audience signals, story/copy work, team roles, and risks.

What it demonstrates:

  • a map that feels like two months of real work
  • multiple owners across one live production surface
  • suggested filters and deep links embedded in the document
  • how ids, tags, metadata, and typed cross-links can support a dense operating map
  • typed cross-links and backlinks across roadmap, docs, launch, and team branches

Good first jumps:

mdm view examples/lantern-studio-map.md#lantern/execution/now
mdm view examples/lantern-studio-map.md#lantern/timeline/week-8

Good CLI probes:

mdm find examples/lantern-studio-map.md "@owner:mira" --plain
mdm find examples/lantern-studio-map.md "#blocked" --plain
mdm kv examples/lantern-studio-map.md --keys owner,area,priority,dependency --plain
mdm links examples/lantern-studio-map.md --plain
mdm relations examples/lantern-studio-map.md#lantern/product/docs --plain

game-world-moonwake.md

A narrative exploration game map with regions, factions, characters, quests, systems, production work, and playtest feedback.

What it demonstrates:

  • worldbuilding that stays operational
  • region metadata like @region:*
  • quest arcs and production tasks in one structure
  • filtering creative work by owner, region, or quest state

Good first jumps:

mdm view examples/game-world-moonwake.md#moonwake/world
mdm view examples/game-world-moonwake.md#moonwake/quests/main
mdm view examples/game-world-moonwake.md#moonwake/production/current

Good CLI probes:

mdm find examples/game-world-moonwake.md "#quest @status:active" --plain
mdm kv examples/game-world-moonwake.md --keys owner,region,status --plain
mdm find examples/game-world-moonwake.md "@region:glass-marsh" --plain

novel-research-writing-map.md

A novel-development workspace with characters, places, plot lines, secondary themes, image systems, quotes, chapter planning, research notes, and revision work.

What it demonstrates:

  • writing projects as structured maps instead of loose notes
  • #theme, #quote, #chapter, and location-based metadata
  • cross-links between characters, locations, themes, plot lines, and current chapters
  • attached detail lines for character and location notes that are longer than a one-line label
  • how research, drafting, and revision can live in one file

Good first jumps:

mdm view examples/novel-research-writing-map.md#glass-archive/themes
mdm view examples/novel-research-writing-map.md#glass-archive/chapters

Good CLI probes:

mdm tags examples/novel-research-writing-map.md --plain
mdm find examples/novel-research-writing-map.md "#quote" --plain
mdm find examples/novel-research-writing-map.md "@location:archive" --plain
mdm kv examples/novel-research-writing-map.md --keys pov,location,owner,status --plain
mdm relations examples/novel-research-writing-map.md#glass-archive/characters/mara --plain

Suggested Workflow

If you are trying to learn the product quickly:

  1. Open one example in mdmind.
  2. Use : or Ctrl+P to jump by branch name, id, #tag, or @metadata.
  3. Use / and b to compare freeform queries with browse-driven discovery.
  4. Use m after narrowing the working set to navigate the same scope in the spatial canvas.
  5. Use M if you want the older fixed visual mindmap for a static read or PNG export.
  6. Use the same file in mdm with find, tags, kv, and links to see the read-only CLI model.