build-from-issue

Given a GitHub issue number, plan and implement the work described in the issue. Operates iteratively - creates an implementation plan, responds to feedback, and only builds when the 'state:agent-ready' label is applied. Includes tests, documentation updates, and PR creation. Trigger keywords - build from issue, implement issue, work on issue, build issue, start issue.

Quality

88%

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

77%Weight 40%Scale 1-3

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

A highly actionable, well-sequenced workflow with strong validation and feedback loops, but it is long and monolithic with some redundant command duplication. Splitting templates and references into bundle files and trimming the redundant command table would improve token efficiency and navigation.

Suggestions

Remove the 'Useful Commands Reference' table — every command already appears inline in its relevant step, so the table is redundant duplication that costs tokens.

Extract the plan-comment, PR-body, and E2E-attestation heredoc templates into a references/ file (e.g. templates.md) and link to it, keeping SKILL.md a leaner overview of the state machine and steps.

State each agent comment marker once in the 'Agent Comment Markers' section and reference it from the heredoc examples instead of repeating the full marker line verbatim in every template.

Dimension	Reasoning	Score
Conciseness	The body is mostly procedural and lean, but the 'Useful Commands Reference' table duplicates commands already shown inline in their relevant steps, and the agent comment markers are restated verbatim inside every heredoc example.	2 / 3
Actionability	Provides fully executable gh/git commands, copy-paste-ready heredoc templates for plan/PR/attestation comments, and concrete worked example scenarios — instructions rather than descriptions.	3 / 3
Workflow Clarity	An explicit state-machine decision tree, branches A–D, numbered Steps 1–14, retry loops (up to 3 attempts per phase), conflict detection, and explicit STOP points give clear sequencing with validation checkpoints.	3 / 3
Progressive Disclosure	Well-organized with sections, but it is a monolithic ~710-line single file with no bundle/reference files; the plan, PR, and attestation templates and the command table are inline content that could be split out.	2 / 3
	Total	10 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 strong, specific description with explicit trigger guidance and concrete actions. It clearly communicates both what the skill does and when to invoke it, with little risk of conflicting with other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'creates an implementation plan, responds to feedback', 'only builds when the state:agent-ready label is applied', and 'Includes tests, documentation updates, and PR creation' — rather than vague language.	3 / 3
Completeness	States both what the skill does (plan and implement GitHub issue work iteratively) and when to use it via an explicit 'Trigger keywords' clause, which is equivalent to a 'Use when...' trigger.	3 / 3
Trigger Term Quality	Provides five natural trigger phrases a user would plausibly say ('build from issue, implement issue, work on issue, build issue, start issue'), giving good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (GitHub issue → plan/build gated by a human-applied label) with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (716 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: NVIDIA/OpenShell
Path: .agents/skills/build-from-issue/SKILL.md
Commit: deced87

Reviewed: 6 days 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.