CtrlK
BlogDocsLog inGet started
Tessl Logo

openspec-explore

Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.

81

1.04x
Quality

73%

Does it follow best practices?

Impact

94%

1.04x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.github/skills/openspec-explore/SKILL.md

The canonical home for this skill is openspec-explore in arm/mlia

SKILL.md
Quality
Evals
Security

Quality

Content

67%Weight 40%Scale 1-5

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

The body is a well-organized, actionable stance guide with concrete commands and artifact references rather than abstract prose. Its main weakness is token efficiency — lengthy ASCII diagrams and repeated example transcripts pad a skill whose core guidance is comparatively compact.

Suggestions

Collapse the multi-line ASCII diagram blocks ("What You Might Do", "Handling Different Entry Points") into shorter representative sketches or move several example transcripts into a referenced file to cut token cost.

Consolidate overlapping guidance — "The Stance", "What You Might Do", "Guardrails", and "What You Don't Have To Do" restate the same principles; merge into a single concise list.

Add one explicit end-of-session validation checkpoint (e.g. 'confirm the user wants to capture before writing any artifact') to make the auto-capture guardrail a concrete sequence step.

DimensionReasoningScore

Conciseness

The prose is mostly efficient and avoids explaining basics Claude already knows, but the large ASCII diagram blocks and several long example transcripts repeat the same stance guidance and could be trimmed without losing clarity.

3 / 5

Actionability

Concrete commands (`openspec list --json`), explicit artifact paths (proposal.md, design.md, tasks.md), and example offer phrases give actionable guidance, with only minor gaps in coverage of edge cases.

4 / 5

Workflow Clarity

Though explicitly a stance rather than a fixed workflow, the change-exists case is sequenced (read artifacts, reference naturally, offer to capture, user decides); as a read-only thinking skill it needs no destructive-operation validation checkpoints.

4 / 5

Progressive Disclosure

Content is well sectioned with clear headers and one-level references to OpenSpec artifact paths; no nested references. Some long inline example transcripts could arguably live elsewhere, but the skill stays navigable at ~280 lines with no bundle files.

4 / 5

Total

15

/

20

Passed

Description

78%Weight 40%Scale 1-5

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

A clear, third-person description that names its domain, several concrete activities, and an explicit 'Use when' trigger. It is well differentiated from implementation skills and reads naturally, with only minor room for broader trigger-term coverage.

DimensionReasoningScore

Specificity

The description lists several concrete actions — "exploring ideas, investigating problems, and clarifying requirements" — giving clear capability coverage, though it stops short of an exhaustive action list.

4 / 5

Completeness

It states both what the skill is ("a thinking partner for exploring ideas...") and when to use it ("Use when the user wants to think through something before or during a change"), with the 'when' clause explicit but somewhat general.

4 / 5

Trigger Term Quality

Natural phrases like "think through something" and "exploring ideas" mirror how a user would phrase the need, but synonyms and common variations are limited.

4 / 5

Distinctiveness Conflict Risk

The "explore mode" thinking-partner framing carves a clear niche with distinct triggers and minimal overlap with implementation-oriented skills.

5 / 5

Total

17

/

20

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
arm/mlia
Reviewed

Table of Contents

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.