create-spike

Investigate a plain-language problem description by deeply exploring the codebase, then create a structured GitHub issue with technical findings. Prequel to build-from-issue — maps vague ideas to concrete, buildable issues. Trigger keywords - spike, investigate, explore, research issue, technical investigation, create spike, new spike, feasibility, codebase exploration.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

71%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 skill body is highly actionable with a complete, copy-paste-ready issue template and a clear sequenced workflow with disposition checkpoints. Its main weaknesses are repetition across the worked examples and a monolithic structure that inlines a large template rather than offloading it to a reference file.

Suggestions

Move the ~80-line issue body template into a references file (e.g. references/issue-template.md) and link to it from Step 4, reducing inline bulk and improving progressive_disclosure.

Collapse the three near-identical example walkthroughs into one detailed example plus a compact table of variations, cutting redundant tokens and lifting conciseness.

Remove the 'Design Principles' restatements of rules already stated inline in Steps 1–5, or compress them to a short checklist, to trim padding.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with specific guidance, but the three full example walkthroughs (feature, bug, performance) repeat the same step pattern, and the 'Design Principles' section restates rules already established earlier (no follow-up comments, one round of clarification), so it could be tightened — matching the 3 anchor rather than the lean 4 or 5.	3 / 5
Actionability	Fully executable, copy-paste-ready guidance: a complete `gh issue create` heredoc with the full issue body template, exact `gh label list` commands, and a concrete checklist for the `principal-engineer-reviewer` sub-agent prompt, covering the common cases per the 5 anchor.	5 / 5
Workflow Clarity	A clear Step 1–5 sequence with an explicit disposition-readiness checkpoint (state:validated vs state:needs-info) and a clarification policy; minor validation gaps (no explicit confirmation the issue was created beyond displaying the URL) keep it just below the 5 anchor's explicit feedback loops.	4 / 5
Progressive Disclosure	Content is well-sectioned with clear headers, but there are no bundle files in references/scripts/assets and the large issue-body template (~80 lines) plus three long examples are inlined rather than split into separate files, fitting the 3 anchor where content that should be separate is inline.	3 / 5
	Total	15 / 20 Passed

Description

91%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.

The description is strong: it states concrete actions, supplies a rich set of natural trigger keywords, and explicitly answers both what and when in third person. The only soft spot is that several of the trigger terms are broad enough to invite minor overlap with other investigation-style skills.

Dimension	Reasoning	Score
Specificity	Lists several specific concrete actions — 'deeply exploring the codebase', 'create a structured GitHub issue with technical findings', 'maps vague ideas to concrete, buildable issues' — with only minor gaps in coverage, fitting the 4 anchor better than the fully comprehensive 5.	4 / 5
Completeness	Explicitly answers both what (investigate codebase and create a structured GitHub issue with technical findings) and when (concrete 'Trigger keywords' phrase), matching the 5 anchor; it is clearly above the 4 anchor where 'when' is only weakly implied.	5 / 5
Trigger Term Quality	Comprehensive coverage of natural terms including synonyms and variations — 'spike, investigate, explore, research issue, technical investigation, create spike, new spike, feasibility, codebase exploration' — matching the 5 anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche — mapping vague ideas to buildable GitHub issues as 'Prequel to build-from-issue' — but the broad terms 'investigate', 'explore', and 'research' carry minor overlap risk with general exploration skills, placing it just below the minimal-conflict 5 anchor.	4 / 5
	Total	18 / 20 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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: NVIDIA/OpenShell
Path: .agents/skills/create-spike/SKILL.md
Commit: 69a05eb

Reviewed: 1 day 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.