CtrlK
BlogDocsLog inGet started
Tessl Logo

ainativedev/aidevcon-2026-ldn

AI Native DevCon 2026 London — all conference sessions as interactive skills

71

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Overview
Quality
Evals
Security
Files

Quality

Content

85%

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

This is a well-crafted skill that provides clear, actionable workflows for answering questions about a specific conference talk. Its strengths are the explicit grounding rules that prevent hallucination, the well-structured progressive disclosure across supporting files, and the multiple use-case sections that cover different query types. The only minor weakness is some repetitiveness in referencing the core workflow across sections, though this serves a reinforcement purpose.

DimensionReasoningScore

Conciseness

The content is mostly efficient but has some redundancy — the grounding rules are repeated implicitly across multiple sections (e.g., 'Follow the workflow above' appears four times). Some phrasing could be tightened, but it generally avoids explaining things Claude already knows.

2 / 3

Actionability

The skill provides highly concrete, step-by-step instructions for every scenario: how to answer factual questions, how to apply concepts, how to teach, and when to proactively surface the talk. Each workflow has specific, numbered steps with clear decision points (e.g., 'if her approach doesn't fit, say so').

3 / 3

Workflow Clarity

The grounding rules establish a clear 5-step workflow with validation (check outline first, then transcript, use verbatim quotes, cite line numbers, acknowledge gaps). Each use-case section builds on this workflow with additional specific steps. The instruction to say 'the talk doesn't address this' when claims aren't found serves as an explicit validation checkpoint.

3 / 3

Progressive Disclosure

The skill is well-structured as an overview that clearly references three supporting files (outline.md, transcript.md, quotes.md) with specific descriptions of what each contains and when to consult them. References are one level deep and clearly signaled. The content is appropriately split between the skill body (workflow/rules) and the supporting files (content/data).

3 / 3

Total

11

/

12

Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This is a strong, well-crafted description that clearly identifies a narrow niche (a specific conference talk by a named speaker) and enumerates the concrete subtopics it covers. It opens with an explicit 'Use when' clause and includes rich, natural trigger terms that a user would plausibly use. The only minor weakness is that it's a single long sentence, but the content is excellent and highly distinctive.

DimensionReasoningScore

Specificity

Lists multiple specific concrete topics: ~350k-line Rust S3 clone experiment, test oracles, flaky tests with AI agents, 100% test coverage critique, human-in-the-loop AI coding, AI-assisted performance engineering, type system invariants, tracing as debugging tool, brownfield/legacy modernisation.

3 / 3

Completeness

Explicitly answers both 'what' (covers Katie Roberts's talk content across many specific subtopics) and 'when' (opens with 'Use when the user asks about...' providing clear trigger guidance).

3 / 3

Trigger Term Quality

Includes highly specific natural trigger terms a user would say: the speaker's name 'Katie Roberts', the exact talk title, plus domain terms like 'brownfield', 'legacy modernisation', 'flaky tests', 'test oracles', 'human-in-the-loop', 'Rust S3 clone', 'tracing', 'type system'. These are terms someone asking about this talk would naturally use.

3 / 3

Distinctiveness Conflict Risk

Extremely distinctive — tied to a specific person's specific talk with a unique title and very particular subtopics (e.g., ~350k-line Rust S3 clone). Virtually no risk of conflicting with other skills.

3 / 3

Total

12

/

12

Passed

Validation

81%

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

Validation9 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

9

/

11

Passed

Reviewed

Table of Contents