Conciseness

The body is mostly efficient, but the "Why These Decisions Are Hard" section explains motivation Claude already understands, and "The Avoiding-It Test" section repeats the cost-of-delay point already made in Step 1's aftermath prose ("The longer you avoid a hard call, the worse the situation usually gets" vs. "Every month you wait, the problem compounds"). It is not a 2 because the core framework steps themselves are tight and free of padding; it is not a 4 because the duplicated avoidance messaging is a real trim opportunity.

Actionability

Concrete, executable guidance throughout: three specific 10/10/10 questions, the exact Andy Grove test question, a fill-in stakeholder impact table, five named communication-plan items, and per-decision question checklists with "The rule" heuristics. Not a 5 because some guidance remains abstract ("does that sentence ring true?"), the empty stakeholder table has no instructions on how to use the filled result, and the description's promised regret-minimization pass has no corresponding step in the body.

Workflow Clarity

Six clearly sequenced steps with go/no-go checkpoints embedded (the Reversibility Test raises the certainty bar, the Pre-Announcement Test gates readiness, the Avoiding-It Test is a checklist). Not a 5 because the relationship between the six-step sequence and the decision-specific frameworks sections is never stated (does the user run the steps then the specific questions?), and Step 4's table output feeds into no explicit next action. Not a 3 because validation gates are explicit, not merely implicit.

Progressive Disclosure

A single ~150-line file with well-organized headers and clearly signaled references ("See ../executive-mentor/references/hard_things.md — Co-Founder Conflicts for full framework"). Not a 5 because both references point outside this skill's bundle to a sibling skill's file, creating a fragile dependency that cannot be verified from the skill itself, and the decision-specific frameworks (co-founder, layoffs, pivot, product kill) are natural candidates for one-level-deep reference files. Not a 3 because what is present is well-signaled and appropriately concise.

Specificity

The description names the domain ("Framework for decisions with no good options") and 1-2 concrete actions ("structured 10/10/10 + regret-minimization pass"), but is not comprehensive — the body's six-step structure, stakeholder mapping, and communication planning are not surfaced. It is not a 4 because only two specific methods are named; it is above a 2 because those methods are genuinely concrete rather than generic.

Completeness

Both what and when are explicitly answered with concrete trigger phrases: the what is "Framework for decisions with no good options... structured 10/10/10 + regret-minimization pass", and the when is an explicit "Use when every option is painful... e.g. choosing between a layoff and a down round, or killing a beloved product line". This matches the top anchor; it is not a 4 because the when-clause is already explicit and illustrated with specific scenarios rather than vague.

Trigger Term Quality

Good natural keyword coverage: "every option is painful", "layoff", "down round", "killing a beloved product line", "10/10/10", "regret-minimization" are phrases a founder or executive would plausibly say. A few natural terms are missing (e.g. "tough decision", "letting someone go", "shut down", "pivot"), keeping it below a 5.

Distinctiveness Conflict Risk

Clear niche (irreversible executive decisions) with distinct triggers (10/10/10, regret-minimization, layoff vs. down round) that no generic decision-making or productivity skill would claim. Minimal conflict risk; matches the top anchor.