x-do

Use when the user asks to build, implement, fix, or execute a plan — detects context (existing plan, new feature, bug, quick task, visual input) and routes through brainstorming, planning, debugging, or execution workflows

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

62%

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

A well-structured, strongly sequenced router skill with clear validation gates and feedback loops, hampered by verbosity from repeated procedural machinery and by several inline references to bundle files (gotchas.md, steps/step-*.md) that do not exist in the provided bundle.

Suggestions

Add the missing referenced files (gotchas.md, steps/step-01-gather.md, steps/step-04-execute.md) or remove/inline their citations so no body pointer is broken.

Collapse the repeated 'x-do routes, it does not execute' rule into a single Self-check stanza and trim the ISOLATE_APPLIED/worktree sub-steps into a referenced file to reduce body length.

Pull the most mode-critical routing decisions out of references/mode-guidance.md into a compact inline table so the body is actionable without a round-trip to a reference.

Dimension	Reasoning	Score
Conciseness	The skill assumes Claude's competence (no basic-concept explanations) and uses terse imperative prose, but at ~196 lines it is dense with procedural machinery — slot-resolution cascades, worktree/--wt flag parsing, ISOLATE_APPLIED states, DOCKER CONTEXT reconstruction — and repeats its core rule ('x-do routes. It does not execute.') across the Role, Mode D, and Self-check sections; could be tightened without loss.	2 / 3
Actionability	Concrete dispatch commands and detection signals are present ('Skill: x-skills:x-review <target>', 'npx tsc --noEmit', 'npx eslint <changed-files>', the 4-mode detection table), but a large share of guidance is deferred to references ('See references/available-tools.md for the full tool table', 'See references/mode-guidance.md for detailed per-mode instructions') and to missing steps/ files, leaving the body itself incomplete rather than copy-paste ready.	2 / 3
Workflow Clarity	A clearly sequenced pipeline (Bootstrap → Research Gate → Detection → Pre-Flight → Routing Signals → Mode Guidance → Implementation Discipline → Post-Implementation Verification → Commit Recomposition → After) with explicit MANDATORY checkpoints and feedback loops (verification-failed → re-review, test-failed → x-bugfix with retries, 'Fix all errors before proceeding').	3 / 3
Progressive Disclosure	Existing references are well-signaled one level deep ('See references/available-tools.md', mode-guidance, delegation-and-scaling, omo-routing), but the body also cites gotchas.md, steps/step-01-gather.md, and steps/step-04-execute.md which are NOT present in the bundle, plus many ../x-shared/*.md external files — broken/absent referenced paths prevent a level-3 score.	2 / 3
	Total	9 / 12 Passed

Description

92%

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 with an explicit trigger clause and concrete enumerated actions and contexts. Its main weakness is overlap with the very skills it routes to, since 'fix' and 'plan' are also primary triggers for x-bugfix and writing-plans.

Suggestions

Add a scoping qualifier to reduce conflict with the executors it dispatches to, e.g. 'Use as the single entry point when the user asks to build, implement, fix, or execute a plan and the right workflow is unclear — routes to …'.

Lead with the router/entry-point identity so Claude picks this over a specialized skill when intent is ambiguous.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and contexts — 'build, implement, fix, or execute a plan', 'detects context (existing plan, new feature, bug, quick task, visual input)', 'routes through brainstorming, planning, debugging, or execution workflows' — matching the 'lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers both 'what' (detects context and routes through workflows) and 'when' (Use when the user asks to build/implement/fix/execute), with an explicit 'Use when...' trigger clause present.	3 / 3
Trigger Term Quality	Opens with natural user phrasing 'Use when the user asks to build, implement, fix, or execute a plan' — these are terms a user would actually say, giving good coverage of natural trigger terms.	3 / 3
Distinctiveness Conflict Risk	The routing/detection framing is distinctive, but the trigger verbs 'build, implement, fix' overlap heavily with sibling skills it dispatches to (x-bugfix for 'fix', writing-plans for 'plan'), so it could fire for the wrong specialized skill; not a level 3 clear niche, but more specific than level 1 generic.	2 / 3
	Total	11 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: quangtran88/x-skills
Commit: 6381b15

Reviewed: 2 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.