decompose-research-goal

SkillDev tools

Lets your agent break a research goal into a tree of smaller testable sub-goals with dependencies and feasibility notes.

Available today. Use it from your connected AI after setup.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the decompose-research-goal skill

About this skill

Turn a research direction into an AND/OR goal DAG with testable leaves, dependency structure, and feasibility annotations.

What this skill tells your AI

The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/decompose-research-goal/SKILL.md and read by ahel’s review.

Purpose

Turn a research direction into an AND/OR goal DAG with testable leaves, dependencies, and feasibility annotations.

Input contract

required: [research_direction, constraints]
optional: [actor_profile, obstacle_report, timeline, north_star]
constraints: [goal wording, constraints, and source context must be explicit]

Execution protocol

Do not perform called SOP operations inline; each loaded SOP owns its contract and thresholds.

  1. You MUST load skill formulate-top-goal to formalize the top goal and desired effect.
  2. You MUST load skill decompose-and-or-goal to recursively construct AND/OR branches and executable leaves.
  3. You MUST load skill validate-goal-tree to check leaf specificity, testability, coverage, and dependency consistency.
  4. You MUST load skill assess-goal-feasibility to annotate feasibility and alternatives.
  5. You MUST load skill map-dependencies to map the goal dependencies. You MUST load skill crystallize-north-star to compress the result into a North Star.

Deviation: Omit feasibility annotations when no actor, resource, or timeline constraints are supplied; mark the tree provisional rather than inventing constraints. Re-run only the affected branch after a decision changes its parent goal.

Output contract

produces: [goal_tree_dag, executable_leaves, dependency_map, feasibility_annotations, research_north_star]
delta_fields: [findings, hypothesis_updates, decisions, uncertainties, open_questions]

Thresholds and quality gates

  • Every leaf must be testable or executable and trace to a parent goal.
  • AND/OR semantics must be explicit; unresolved feasibility remains an uncertainty.
  • Completion requires coverage of the declared top goal, not a fixed leaf count.

Failure and counterexamples

Reject trees with circular dependencies, vague leaves, or branches that cannot be tied to the declared direction. Do not treat a polished sentence as evidence that decomposition is complete.

Provenance map

  • resolved: goal-decomposition
  • resolved: north-star-crystallization-formulate-top-goal
  • resolved: north-star-crystallization-and-or-decompose
  • resolved: north-star-crystallization-validate-leaves
  • resolved: north-star-crystallization-feasibility-check
  • resolved: north-star-crystallization-ask-decomposition-validation

Preserved source criteria ledger

sourcesource linekindsource criterion
none retained--No source numeric/textual criterion retained after normalization.

Context checkpoint / Delta notes

Append the goal DAG, leaf validation, feasibility annotations, dependency changes, and unresolved validation questions.

Signals

GitHub stars
501
Forks
41
Last commit
Sep 2026
Advanced
Catalog kind
skill
Key
decompose-research-goal
Source
github.com/yogsoth-ai/de-anthropocentric-research-engine