Model product UI actions from user intent questions after a meaningful outcome, completion, created resource, imported data, uploaded file, synced integration, report, deployment, automation, review state, handoff, or workflow milestone. Use when deciding which actions to show, hide, group, name, prioritize, defer, or place across outcome, item, selection, continuation, navigation, recovery, and assistance scopes.
44
46%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/payoff-action-modeling/SKILL.mdQuality
Discovery
52%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is structurally complete with both 'what' and 'when' clauses, but suffers from dense, jargon-heavy language that would be difficult for Claude to match against natural user requests. The enumeration of scopes and milestone types reads like an internal taxonomy rather than a user-facing description, significantly reducing its practical effectiveness for skill selection.
Suggestions
Replace jargon-heavy scope names ('outcome, item, selection, continuation, navigation, recovery, assistance scopes') with plain-language examples of what users would actually ask, e.g., 'Use when a user asks what actions or buttons to show after completing a task, uploading a file, or finishing onboarding.'
Add natural trigger terms that users would actually say, such as 'next steps UI', 'post-action buttons', 'what to show after signup', 'empty state actions', or 'contextual actions'.
Simplify the dense enumeration into a shorter, clearer description with 2-3 concrete examples rather than exhaustively listing every possible milestone and scope type.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names a domain (product UI actions) and lists many actions (show, hide, group, name, prioritize, defer, place), but the language is dense and abstract rather than concretely illustrative. It reads more like a taxonomy than a clear capability list. | 2 / 3 |
Completeness | The description does explicitly answer both 'what' (model product UI actions from user intent questions after milestones) and 'when' ('Use when deciding which actions to show, hide, group, name, prioritize, defer, or place across...scopes'). The 'Use when' clause is present and detailed. | 3 / 3 |
Trigger Term Quality | The terms used are highly specialized jargon ('outcome scopes', 'continuation scopes', 'handoff', 'workflow milestone', 'assistance scopes') that users would rarely naturally say. A user needing help with UI action design would more likely say things like 'what buttons to show after signup' or 'next steps UI', none of which are represented here. | 1 / 3 |
Distinctiveness Conflict Risk | The description is fairly niche in its focus on post-milestone UI action modeling, which reduces conflict risk. However, the dense enumeration of scopes and triggers is so broad within the UI/UX domain that it could overlap with general UI design or UX pattern skills. | 2 / 3 |
Total | 8 / 12 Passed |
Implementation
39%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a thorough and well-structured framework for modeling post-outcome UI actions, with a clear 8-step workflow and useful reference tables. However, it is severely over-length and verbose, repeating concepts across sections (scope definitions, action types) and including exhaustive lists that should be in separate reference files. The lack of progressive disclosure and the excessive token cost significantly undermine its effectiveness as a SKILL.md.
Suggestions
Reduce the SKILL.md to a concise overview (~80-100 lines) covering the core output format, outcome state types, key principles (briefly), and the 8-step workflow, then move the 44-item question bank, anti-patterns list, action placement rules, and common outcome examples into separate referenced files (e.g., QUESTION_BANK.md, ANTI_PATTERNS.md, PLACEMENT_RULES.md).
Eliminate redundant definitions — action scopes are defined in 'Core Output', re-explained in 'Workflow step 3', and referenced again in the review checklist. Define once and reference.
Add one complete worked example showing the full workflow applied to a specific product scenario (e.g., 'AI generates 15 test cases') with the final action model table filled out, so Claude has a concrete reference output.
Remove explanatory prose that restates obvious UI design principles (e.g., 'Actions should appear near the object they affect' and 'Avoid overwhelming users immediately after success') — Claude already knows these concepts and they consume significant tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines. It extensively explains UI design concepts (locality principle, intent pressure, action density) that Claude already understands. The question bank alone is 44 items, the anti-patterns list is 23 items, and much content is redundant across sections (e.g., scope definitions appear in multiple places). This could be compressed to roughly one-third its size without losing actionable value. | 1 / 3 |
Actionability | The skill provides concrete table formats, a structured workflow (steps 1-8), specific label recommendations, and a common outcomes table with examples. However, there is no executable code or copy-paste-ready output template — the table shape is described but not shown as a complete filled example for a realistic scenario. The guidance is specific but remains at the level of structured instructions rather than fully worked examples. | 2 / 3 |
Workflow Clarity | The 8-step workflow is clearly sequenced with explicit substeps, and step 8 is a dedicated review/validation checkpoint with a comprehensive checklist. The process flows logically from defining the outcome through classification, placement, deduplication, and final review. For a design-oriented (non-destructive) skill, this level of workflow structure with a review checklist is strong. | 3 / 3 |
Progressive Disclosure | The entire skill is a monolithic wall of text with no references to external files or bundle resources. Content that could be split out — the 44-item question bank, the anti-patterns list, the action placement rules, the common outcome examples — is all inline. There is no layered structure; everything is presented at the same level in a single file. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
64f57c5
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.