railway-test

Triggered when the user invokes /railway-test or $railway-test, asks to test a pull request on Railway, or requests browser-based exact-head preview QA and pull-request evidence before merging.

Quality

84%

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

81%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.

A highly actionable, well-sequenced workflow with strong validation checkpoints and clean progressive disclosure to real bundle files. The main weakness is repetition of the same bearer/head-SHA guardrails across multiple steps rather than stating them once.

Suggestions

Consolidate the repeated bearer/secrecy and head-SHA re-read guardrails into a single early "Guardrails" section referenced from each step, instead of restating them in steps 2, 3, and 7.

Move the long comment-reconciliation sub-procedure (paginate/PATCH/DELETE) into a scripts/ helper or references/ file so the body stays a lean overview.

Trim redundant restatements of the status-gate rules between step 1's "Mandatory status gate" and step 7's evidence derivation.

Dimension	Reasoning	Score
Conciseness	The body is dense with specific operational guidance and avoids explaining concepts Claude already knows, but it repeats safety refrains (bearer handling, head-SHA re-reads) several times across steps, which could be tightened into a single shared guardrail.	3 / 5
Actionability	Provides fully executable, copy-paste-ready commands throughout (gh pr view/diff, preview_state.sh, paginated comment listing, PATCH/DELETE reconciliation, browser.tabs.finalize) covering the common cases concretely.	5 / 5
Workflow Clarity	A clearly sequenced 7-step workflow with explicit validation checkpoints (mandatory status gate, build polling, head-SHA re-reads before opening and publishing, reconcile-to-exactly-one-comment) and feedback loops for the destructive/external PR-comment operation.	5 / 5
Progressive Disclosure	Body points to real one-level-deep bundle files (references/test-recipes.md, references/streaming-cadence.md, scripts/preview_state.sh) that are clearly signaled inline, though the core body itself is fairly long and could offload more detail into references.	4 / 5
	Total	17 / 20 Passed

Description

87%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 strong, third-person description that clearly states both capability and explicit trigger conditions with low conflict risk. The only weakness is that the concrete actions are bundled into a clause rather than enumerated, costing a touch of specificity.

Dimension	Reasoning	Score
Specificity	Names several concrete actions — "test a pull request on Railway", "browser-based exact-head preview QA", "pull-request evidence before merging" — but they are somewhat bundled rather than enumerated, leaving minor coverage gaps versus a comprehensive list.	4 / 5
Completeness	Explicitly answers both what (test a PR on Railway, browser-based exact-head preview QA, PR evidence) and when via the explicit "Triggered when the user invokes... or asks... or requests..." clause with concrete trigger phrases.	5 / 5
Trigger Term Quality	Good coverage of natural phrases users would say ("invokes /railway-test", "test a pull request on Railway", "preview QA"), with a few synonyms or variant phrasings missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (exact-head Railway preview QA with PR evidence) with distinct, specific triggers and minimal realistic overlap with other skills.	5 / 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: nearai/ironclaw
Path: .claude/skills/railway-test/SKILL.md
Commit: b0b999d

Reviewed: 5 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.