talk-sloan-harness-engineering-beyond-code

Summarizes Rob Sloan's harness-engineering talk and creates safe design artifacts for agent context beyond code: product-intent packets, design constraints, acceptance criteria, context ownership, and review gates. Use when the user asks about making non-code context agent-ready or improving AI work with product/design intent.

Quality

85%

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

82%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 concise, well-structured, and actionable for an instruction-only reference skill, with clear workflows and an explicit redaction branch. Its main gap is progressive disclosure: the read order points to outline.md/quote.md/transcript.md, but no bundle files are present, leaving those references dangling.

Suggestions

Add the referenced bundle files (outline.md, quote.md, transcript.md) under references/ so the read-order navigation resolves, or remove the file-specific read order and inline the essential content.

Add an explicit validation/review step to the Core Workflow (e.g., 'Confirm each acceptance criterion is testable before declaring the artifact done') to turn the implicit review-gate guidance into a concrete checkpoint.

Tighten abstract steps like 'Answer in 2-5 sentences' into a concrete expected output shape so the guidance is closer to copy-paste-ready.

Dimension	Reasoning	Score
Conciseness	The body is lean — short read-order list, compact numbered workflows, and terse output templates with no padded concept explanations, assuming Claude's competence; every section earns its place, matching the score-5 anchor.	5 / 5
Actionability	Concrete guidance is present via numbered workflows and explicit output templates (Goal/Boundaries/Review points/Evidence/Open questions) plus response-shape examples, but some steps stay abstract ("Answer in 2-5 sentences", "State the product intent in one paragraph"), so it sits above the score-3 anchor yet short of fully copy-paste-ready score-5 guidance.	4 / 5
Workflow Clarity	Two clear numbered sequences (factual Q&A and applying-the-talk) plus an explicit redaction branch ("If the user asks for omitted mechanics... answer with the safe design principle") give a clear sequence with an error/edge-case checkpoint, but explicit validation of produced artifacts is implicit rather than stated, capping it just below score 5.	4 / 5
Progressive Disclosure	The body uses well-signaled, one-level-deep references with a numbered read order (outline.md, quote.md, transcript.md) and a clear overview, matching the score-4 anchor; it is not a 5 because the referenced bundle files do not actually exist in references/, so navigation is designed well but the referenced paths are missing.	4 / 5
	Total	17 / 20 Passed

Description

88%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 specific, third-person, and cleanly answers both what the skill does and when to invoke it, with a distinct niche. Its main weakness is trigger-term breadth — the triggers are natural but lack synonyms or variants a user might say.

Dimension	Reasoning	Score
Specificity	"Summarizes Rob Sloan's harness-engineering talk and creates safe design artifacts" then enumerates product-intent packets, design constraints, acceptance criteria, context ownership, and review gates — multiple concrete actions with comprehensive coverage, matching the score-5 anchor; it is not a 4 because no meaningful actions are missing.	5 / 5
Completeness	It explicitly answers both what ("Summarizes... and creates safe design artifacts...") and when ("Use when the user asks about making non-code context agent-ready or improving AI work..."), matching the score-5 anchor with concrete trigger phrases.	5 / 5
Trigger Term Quality	The "Use when the user asks about making non-code context agent-ready or improving AI work with product/design intent" clause supplies several natural phrases, but lacks common synonyms/variants a user might say, so it sits above the score-3 anchor yet below the comprehensive score-5 anchor.	4 / 5
Distinctiveness Conflict Risk	The harness-engineering/non-code-context niche is mostly distinct with specific triggers, but the broad phrase "improving AI work" creates minor overlap risk with general AI-improvement skills, keeping it just below the score-5 clear-niche anchor.	4 / 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-sloan-harness-engineering-beyond-code/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.