AgentStack
SKILL verified MIT Self-run

Hypothesis Tree

skill-daddia-claude-for-strategy-hypothesis-tree · by daddia

>

No reviews yet
0 installs
3 views
0.0% view→install

Install

$ agentstack add skill-daddia-claude-for-strategy-hypothesis-tree

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Hypothesis Tree? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Hypothesis Tree

When to use

Break a problem into falsifiable root and MECE sub-hypotheses before analysis. See ../../references/hypothesis-driven-approach.md.

What this skill does not do

  • Does not run analysis — names evidence needed; workplan-builder executes.
  • Does not use topic labels as nodes — every node falsifiable.
  • Does not keep branches with no so-what either direction — cut and list.

Preconditions

| Input | If missing | |---|---| | Problem as question | Restate and confirm before tree | | Root hypothesis ambiguity | Present both roots; ask which to build |

Provisional mode

Uncertain root: two candidate roots with user choice; label structured first pass.

Trust spine

  • Analytical Rigor (mandatory): MECE sub-hypotheses; evidence-needed per branch; cut branches listed.
  • Confidence bands (hypothesis-driven-analysis): High = triangulated evidence plan; Medium = single source; Low = assumption-driven — label load-bearing assumption.
  • Per trust-conventions.md for sourcing, numbers, gate.

Workflow

  1. Restate problem as question; confirm.
  2. Propose falsifiable root hypothesis.
  3. 3–5 MECE sub-hypotheses.
  4. Smallest evidence test per sub-hypothesis.
  5. So-what if TRUE and if FALSE; cut if neither matters.
  6. Output as tree.

Output format

ROOT HYPOTHESIS: [falsifiable claim]

1. [Sub-hypothesis 1]
   Evidence needed: [...]
   If TRUE: [...]
   If FALSE: [...]
...

Cut branches: [...]

Worked example

Input: "Why did mid-tier churn spike?"

Excerpt: Root: price increase caused mid-tier churn. Sub-hypothesis: elasticity >15% in segment; evidence: cohort comparison Q1 vs Q4.

Quality checks before delivering

  • [ ] Falsifiable nodes only
  • [ ] MECE check explicit
  • [ ] Cut branches documented
  • [ ] Handoff-ready for workplan-builder

Propose profile update

When a stable convention surfaces during this run (thresholds, naming, tone, output format, or recurring corrections), propose a profile update: show the exact diff against ~/.claude/plugins/config/claude-for-strategy/consulting/CLAUDE.md (org-wide facts go to org-profile.md), ask for confirmation, and write only on yes. Only /consulting:practice-setup auto-applies a full profile write.

Outputs

Follows plugin CLAUDE.md § Outputs. Next: /consulting:workplan-builder.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.