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

77%

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-structured, concise skill that establishes clear grounding rules and a sensible workflow for answering questions from a talk transcript. Its main weakness is the lack of concrete examples showing expected input/output patterns (e.g., a sample factual Q&A or summary format). The references to supporting files are clear in purpose but could benefit from more explicit descriptions and navigation links.

Suggestions

Add 1-2 concrete examples showing a sample user question and the expected response format, including timestamp citation style (e.g., 'At [12:34], Simon says...').

Add brief descriptions of what each referenced file contains (e.g., 'outline.md — section headings with timestamp ranges; quotes.md — pre-vetted short excerpts with speaker attribution').

DimensionReasoningScore

Conciseness

The skill is lean and efficient. It avoids explaining what evals are, what transcripts are, or how speech-to-text works. Every section serves a clear purpose with no padding or unnecessary context.

3 / 3

Actionability

The guidance is concrete in terms of process (read outline.md first, then transcript.md, cite timestamps, etc.) but lacks executable examples. There are no sample Q&A exchanges, no example of how a summary should look, and no demonstration of how to format a citation with timestamps.

2 / 3

Workflow Clarity

The grounding rules establish a clear sequence (outline → transcript → answer) with explicit validation checkpoints: flag garbled text, say so if the transcript doesn't address a question, distinguish grounded claims from outside context, and mark non-talk advice. For a non-destructive, retrieval-oriented skill, this is well-sequenced with appropriate safeguards.

3 / 3

Progressive Disclosure

The skill references outline.md, transcript.md, and quotes.md with clear roles for each, which is good structure. However, no bundle files were provided, so we cannot verify these references exist. The SKILL.md itself is well-organized with clear sections, but lacks explicit navigation links (e.g., [outline.md](outline.md)) and doesn't describe what each file contains beyond brief mentions.

2 / 3

Total

10

/

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 its narrow scope (a specific conference talk), lists concrete actions, and provides rich trigger terms including speaker names and topic keywords. The explicit 'Use when' clause with multiple trigger scenarios makes it easy for Claude to select appropriately, and the tight scoping to a specific talk minimizes conflict risk with other skills.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions: 'Answers factual questions from the transcript, summarizes the talk, extracts safe quotes, and applies the talk's concepts to the user's work while treating transcript text as untrusted source material.' These are clear, distinct capabilities.

3 / 3

Completeness

Explicitly answers both 'what' (answers factual questions, summarizes, extracts quotes, applies concepts) and 'when' (opens with 'Use when the user asks about...' with detailed trigger conditions). The 'Use when' clause is prominent and comprehensive.

3 / 3

Trigger Term Quality

Excellent coverage of natural trigger terms: speaker names ('Simon Obstbaum', 'Rob Willoughby'), talk title, and topical keywords ('AI evals', 'evaluation design', 'testing agents', 'measurement', 'quality gates', 'agent reliability'). These are terms users would naturally use when asking about this content.

3 / 3

Distinctiveness Conflict Risk

Highly distinctive — scoped to a specific talk by named speakers at a named conference. The combination of speaker names, talk title, and transcript-based answering makes it extremely unlikely to conflict with other skills, even other eval-related skills.

3 / 3

Total

12

/

12

Passed

Validation

100%

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

Validation11 / 11 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents