using-superpowers

Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions

Quality

32%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/using-superpowers/SKILL.md

Quality

Discovery

17%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This description functions as a meta/bootstrap skill that fires on every conversation, which makes it inherently problematic as a skill description. It lacks concrete actions, natural trigger terms, and distinctiveness. The description reads more like an internal system instruction than a skill that Claude should selectively choose from a pool of options.

Suggestions

Add specific concrete actions the skill performs, e.g., 'Scans available skill files, indexes their capabilities, and routes user requests to the appropriate skill based on keyword matching.'

Replace the overly broad 'any conversation' trigger with more specific conditions, or if it truly must run on every conversation, clarify what distinguishes it from other skills to reduce conflict risk.

Include natural user-facing keywords that would help Claude understand when this skill adds value beyond other skills, such as 'skill discovery, skill lookup, finding the right tool, capability search.'

Dimension	Reasoning	Score
Specificity	The description does not list any concrete actions or capabilities. It vaguely mentions 'establishes how to find and use skills' and 'Skill tool invocation' without specifying what the skill actually does beyond being a meta-process for skill selection.	1 / 3
Completeness	It has a 'when' clause ('Use when starting any conversation') but the 'what' is extremely weak — it only says it 'establishes how to find and use skills' without explaining what concrete value it provides. The when clause is present but overly broad.	2 / 3
Trigger Term Quality	There are no natural user-facing keywords. Terms like 'Skill tool invocation' and 'establishes how to find and use skills' are internal system jargon that no user would naturally say. The trigger is 'any conversation' which is maximally generic.	1 / 3
Distinctiveness Conflict Risk	Triggering on 'any conversation' means this skill would conflict with every other skill. It has no distinct niche or specific trigger conditions that would differentiate it from other skills.	1 / 3
	Total	5 / 12 Passed

Implementation

47%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill has a clear workflow and good structural organization, but is severely undermined by extreme verbosity and repetition. The core instruction ('always invoke skills before responding') is hammered home through a 12-row table, multiple emphatic sections, and repeated restatements that waste significant token budget. The content would be far more effective at roughly one-third its current length.

Suggestions

Reduce the 'Red Flags' table to 3-4 representative examples instead of 12 variations of the same idea, or replace with a single concise rule statement.

Remove the <EXTREMELY-IMPORTANT> block and SUBAGENT-STOP block — these are motivational/disciplinary framing that Claude doesn't need; the rule itself is sufficient.

Consolidate the repeated 'invoke skills before anything' message into a single clear statement at the top, removing redundant restatements throughout the document.

Dimension	Reasoning	Score
Conciseness	Extremely verbose and repetitive. The core message ('invoke skills before responding') is stated in at least 5 different ways. The 12-row 'Red Flags' table is padded with variations of the same idea. The DOT graph, while structured, adds significant token cost for a simple sequential flow. Much of this content explains things Claude already understands about following instructions.	1 / 3
Actionability	Provides concrete guidance on when and how to invoke skills (tool names per platform, priority order), but the actual actionable content is buried in repetitive motivational/disciplinary language. The DOT graph is a concrete workflow but not directly executable. No concrete code examples since this is an instruction-only skill, but the instructions themselves are moderately specific.	2 / 3
Workflow Clarity	The workflow is clearly sequenced via the DOT digraph: receive message → check for skills → invoke → announce → check for checklist → create todos → follow skill → respond. The priority ordering (process skills first, then implementation) and the instruction priority hierarchy (user > skills > system prompt) are explicit and well-structured.	3 / 3
Progressive Disclosure	References external files (references/copilot-tools.md, references/codex-tools.md) which is good, but the main content is a monolithic wall of text with multiple sections that could be condensed. The skill types section and platform adaptation section feel like they could be separate references. The content is structured with headers but overly long for what it communicates.	2 / 3
	Total	8 / 12 Passed

Validation

100%

Warnings & errors only

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.

Repository: obra/superpowers
Commit: b7a8f76

Reviewed: 33 minutes ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

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.