- Home
- Skills
- Bdambrosio
- Cognitive Workbench
- Mc Map Query
mc-map-query_skill
- Python
9
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill bdambrosio/cognitive_workbench --skill mc-map-query- Skill.md5.0 KB
- tool.py30.0 KB
Overview
This skill queries the agent's persistent spatial map to extract cells that match strategic criteria for planning, safety, and resource tasks. It returns standardized summaries and structured results that support movement planning, hazard assessment, exploration, and resource discovery. Use it to convert the agent's memory into actionable sets of locations.
How this skill works
The skill inspects a cell-based SpatialMap built from past and present observations. You provide a query type and optional parameters (center coordinates, radius, thresholds, resource type, etc.) and the skill filters or computes attributes like reachability, blocking, confidence, and hazard risk. Results are returned as a uniform object with a human-readable summary and an array of matching cells and metadata.
When to use it
- Plan a safe route or pick waypoint candidates near the agent.
- Locate resources (wood, ore, water, food) or specific block instances for gathering.
- Identify unexplored or low-confidence areas for targeted exploration.
- Detect hazards, stale observations, or cells that present drop/climb risks.
- Assess coverage and general map statistics to inform strategy.
Best practices
- Include x,z (and y when relevant) for accurate query-time computations like blocking, reachability, and drop/climb risk.
- Use radius to limit search scope and control performance; defaults are available if omitted.
- Normalize resource_type inputs (minecraft: prefix optional) for exact matches or use generic types for broader results.
- Apply thresholds (confidence, max_age, max_delta_y) to tune sensitivity for stale or low-confidence cells.
- Prefer multi-objective queries (candidate waypoints, nearest) when you need combined safety and reachability checks.
Example use cases
- Find nearby safe tiles to stand on after spotting a hostile mob: query cells-safe-to-stand with center coordinates and radius.
- Target unexplored terrain for efficient mapping: query cells-unobserved or frontier-cells around the agent.
- Plan a climb or descent: query cells-requiring-climb or cells-with-drop-risk providing current y to compute step and drop attributes.
- Gather resources: query cells-with-resource using generic types like 'ore' or a specific ID like 'minecraft:coal_ore'.
- Recover from outdated data: query cells-stale or cells-low-confidence to prioritize re-observation.
FAQ
Generic types (e.g., 'wood', 'ore') match all instances in that category. Specific IDs (e.g., 'minecraft:oak_log') perform exact matches; the 'minecraft:' prefix is optional.
When should I include the y coordinate?
Provide y for queries that compute relative attributes at query time (cells-blocked, cells-reachable, cells-requiring-climb, cells-with-drop-risk) to get accurate step, blocking, and drop-risk results.