CtrlK
BlogDocsLog inGet started
Tessl Logo

x-mindful

Use BEFORE implementation when AI has produced a plan, spec, design doc, PRD, or code-bearing proposal and you want a senior-engineer architect-review pass — walks the user through the decisions, tradeoffs, assumptions, blind spots, shape mismatches, and future-debt AI silently embedded. Surfaces "what AI did, what it picked over alternatives, what it took on, and what gets harder later" one item at a time with confirm / modify / reject / skip gates. Hard-blocks code-level details so the human stays at architectural direction.

67

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Advisory

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Quality

Content

77%

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-structured, highly actionable orchestration overview with a clear phased workflow and explicit verification gate. Its two weaknesses are token padding (the long auto-gate keyword lists) and broken progressive disclosure: the step files it tells the reader to load before each phase are missing from the bundle.

Suggestions

Add the missing steps/step-01-detect.md through step-05-handoff.md (and gotchas.md), or remove the references and inline the essential per-phase instructions — currently the Phase Dispatch table points to files that do not exist.

Trim the auto-gate signal keyword lists (Tradeoff / Boundary / Trust / Operational / Architectural-decision) into a shorter representative set or move the exhaustive list into a reference file to reduce token cost.

De-duplicate the taxonomy, which appears both as prose in the intro paragraph and again as the 'What x-mindful Surfaces' table.

DimensionReasoningScore

Conciseness

The body is competent and non-pedantic (no basic-concept filler, assumes Claude's intelligence) but is long with padding — notably the ~40-keyword auto-gate signal lists across five bullet groups and some taxonomy restatement between the intro and the table — so it could be tightened rather than earning 'every token earns its place'.

2 / 3

Actionability

Provides concrete, copy-paste-ready guidance: an exact decision-envelope markdown template, a precise MCP call signature (mcp__basic-memory__search_notes({ query, page_size: 5 })), a confirm/modify/reject/skip/quit gate menu, and explicit anti-trigger and skip-condition tables — fully actionable for an instruction/orchestration skill.

3 / 3

Workflow Clarity

Sequences five phases in a clear dispatch table, identifies Phase 4 as the core loop with 1-3/5 linear, and includes an explicit MANDATORY Completion/verifier gate with a concrete done-definition and a feedback loop (the gate menu), matching clear sequence with explicit validation steps.

3 / 3

Progressive Disclosure

Signaling and one-level-deep organization are good (References section, lazy-load list, Phase Dispatch table mapping phases to step files), but the core referenced files do not exist: the entire steps/ directory (step-01 through step-05) and gotchas.md are absent, so navigation leads to dead ends; only the four references/*.md files resolve.

2 / 3

Total

10

/

12

Passed

Description

85%

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 highly specific and complete with an explicit Use-when trigger and a distinctive architect-level niche. Its main weakness is trigger-term tone: it reads as professional/technical rather than echoing the natural phrases a user would actually say.

Suggestions

Soften the trigger language toward conversational phrasings a user would say (e.g. add "review this plan", "what could break", "what did AI silently pick") alongside the artifact-type keywords.

Tighten the description — the three long sentences are dense but padded; the confirm/modify/reject/skip enumeration and the quoted 'what AI did...' clause overlap and could be merged without losing the trigger.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "walks the user through the decisions, tradeoffs, assumptions, blind spots, shape mismatches, and future-debt", "confirm / modify / reject / skip gates", "Hard-blocks code-level details" — matching the anchor for listing several specific concrete actions.

3 / 3

Completeness

Explicitly answers both what ("walks the user through...Surfaces...one item at a time with confirm / modify / reject / skip gates") and when ("Use BEFORE implementation when AI has produced a plan, spec, design doc, PRD...") with an explicit trigger clause.

3 / 3

Trigger Term Quality

Includes relevant artifact terms a user would say ("plan, spec, design doc, PRD, or code-bearing proposal", "architect-review") but is more technical than conversational, missing common phrasings like "review this plan", "what could break", or "what did AI miss".

2 / 3

Distinctiveness Conflict Risk

Carves a clear niche — "senior-engineer architect-review pass" that is pre-implementation and "Hard-blocks code-level details so the human stays at architectural direction" — making it unlikely to fire for code review, teaching, or research skills.

3 / 3

Total

11

/

12

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
quangtran88/x-skills
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.