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

100%

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 efficiently defines a clear retrieval-and-citation workflow for answering questions about a specific talk. Its strengths are the tight grounding rules that prevent hallucination, the concrete example showing expected output format with fallback behavior, and the clean progressive disclosure across bundle files. The only minor note is that quotes.md is mentioned in the body but not listed in the bundle file structure section, creating a slight inconsistency.

DimensionReasoningScore

Conciseness

The skill is lean and efficient. Every section serves a clear purpose: grounding rules constrain behavior, the bundle structure section explains dependencies, and the example shows expected output format. No unnecessary explanations of concepts Claude already knows.

3 / 3

Actionability

The skill provides concrete, step-by-step workflows: check outline.md first, then read the relevant section of transcript.md, then answer with verbatim quotes. The example demonstrates exact output format including fallback behavior when quotes aren't available. The grounding rules are specific and enforceable constraints.

3 / 3

Workflow Clarity

The workflow is clearly sequenced: (1) read outline.md to locate section, (2) read that section of transcript.md, (3) answer with verbatim quotes. Validation is built in via the grounding rules (don't speculate, say so if not found). The fallback behavior when files are unavailable is explicitly defined, providing error recovery guidance.

3 / 3

Progressive Disclosure

The skill is well-structured as an overview that references supporting files (outline.md, transcript.md, quotes.md) at one level deep with clear descriptions of each file's purpose. Content is appropriately split — the skill file contains the workflow and rules while detailed content lives in the bundle files. Navigation is clearly signaled.

3 / 3

Total

12

/

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 an excellent skill description that clearly defines a narrow, well-scoped domain (a specific conference talk) with comprehensive trigger terms. It opens with an explicit 'Use when' clause and enumerates numerous specific concepts that would help Claude match user queries accurately. The only minor concern is its length, but the specificity justifies the verbosity in this case.

DimensionReasoningScore

Specificity

The description lists multiple specific concrete concepts: ADR-first workflows with agents, the three pillars (each named), producer 'black box' (harness/host/model), vanity metrics vs adoption, two-to-four-person sub-streams, one-complex-task-at-a-time, and more. These are highly specific and concrete.

3 / 3

Completeness

The description explicitly answers both 'what' (knowledge about Christopher Batey's specific talk and its concepts) and 'when' with a clear 'Use when the user asks about...' clause that enumerates specific trigger scenarios including applying his approach to current work.

3 / 3

Trigger Term Quality

Excellent coverage of natural terms a user might say: the speaker's name 'Christopher Batey', the talk title, specific concepts like 'ADR-first workflows', 'path to production at AI speed', 'AI-enabled engineers', 'you build it, you run it, you drive adoption', and broader terms like 'AI-assisted product engineering teams'. These are terms users would naturally reference.

3 / 3

Distinctiveness Conflict Risk

This is extremely distinctive — it's scoped to a single specific talk by a named speaker at a named conference in a specific year. The trigger terms are highly unique (e.g., 'producer black box', 'harness/host/model', the specific three pillars) and would be very unlikely to conflict 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