talk-thomas-ai-native-engineering

Use when the user asks about Ian Thomas's talk "AI Native Engineering" (Meta / Reality Labs / Horizon Experiences) — including questions about Meta's AI4P (AI For Productivity) programme, the 6-dimension / 5-level AI maturity model and self-assessment workshop, how Horizon rolled out AI tooling across 500+ engineers, engineering excellence as an adoption vehicle, anti-test-slop, autonomous code mods, the DRS risk-scoring tool, the Horizon MCP server, vanity metrics vs real productivity, or applying Thomas's ground-up-plus-top-down adoption playbook to their own org.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%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 body is a well-organized instruction-only reference skill with concrete, sequenced workflows, explicit grounding/safety rules, and clear deferral to referenced files. Its main gaps are a few repeated phrases, no worked answer example, and referenced bundle files that are not present in the bundle.

Suggestions

Bundle the referenced files (outline.md, transcript.md, quote.md) under references/ so the one-level-deep navigation the body promises actually resolves.

Add one short worked example (a sample question and the expected answer shape with a verbatim excerpt and line citation) to make the answer format concrete.

Trim the repeated "safe excerpts" phrasing across sections into a single stated rule referenced once.

Dimension	Reasoning	Score
Conciseness	Largely efficient: grounding rules, safety rules, and concrete procedures without padding about what a transcript or MCP server is. Minor repeats ("safe excerpts" restated across sections, a slightly editorial intro) could be trimmed, keeping it just below lean.	4 / 5
Actionability	Concrete, specific guidance throughout — exact file hops (outline.md -> "Named frameworks / concepts" -> "AI Maturity Model"), exact level names ("sit / walk / jog / run / leap"), and concrete workshop mechanics ("anonymous voting + discussion + repeat every 3-4 weeks"). Lacks a worked example of the expected answer shape, so it stops just short of fully copy-paste-ready.	4 / 5
Workflow Clarity	Each use-case is a numbered, sequenced workflow with explicit checkpoints and fallbacks ("If the answer genuinely isn't in the transcript, say so explicitly"; "ask before scoring"). Validation is mostly disclose-and-stop rather than verify-then-retry loops, so it lands just below the top anchor.	4 / 5
Progressive Disclosure	Well-structured with clear headers and correctly signals one-level-deep references ("check quote.md first", "Use outline.md -> ... to locate"). The referenced files outline.md, transcript.md, and quote.md are not actually present in the bundle (no references/ dir exists), so navigation is not fully clean and it stops short of the top anchor.	4 / 5
	Total	16 / 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.

The description is a strong, explicit trigger-rich statement that names a highly specific niche and enumerates many concrete topics a user would ask about. Its only weakness is that it lists topics rather than the actions Claude performs, slightly capping specificity.

Dimension	Reasoning	Score
Specificity	Lists several specific, distinguishable capabilities and topics ("AI4P programme", "6-dimension / 5-level AI maturity model and self-assessment workshop", "anti-test-slop, autonomous code mods", "DRS risk-scoring tool", "Horizon MCP server"). It enumerates topics rather than the explicit actions Claude performs (answer, apply, audit, surface), keeping it just below comprehensive.	4 / 5
Completeness	Explicitly answers both what (the talk and its enumerated topics) and when, opening with a rich "Use when the user asks about ... including ... or applying ... to their own org" trigger clause with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural user-facing phrasing ("questions about", "applying ... to their own org") plus named entities users would say ("Ian Thomas", "AI Native Engineering", "Meta", "Reality Labs", "Horizon"). Good coverage, though a few synonyms (e.g. plain "AI maturity", "large-scale refactoring") are absent, so it stops short of comprehensive.	4 / 5
Distinctiveness Conflict Risk	The niche is a named individual's named talk at a named org with distinct triggers, giving it a clear, unique slot with minimal realistic overlap against other skills.	5 / 5
	Total	18 / 20 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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: jscraik/Agent-Skills
Path: Plugins/aidevcon/skills/talk-thomas-ai-native-engineering/SKILL.md
Commit: a3808a9

Reviewed: ~3 hours 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.