CtrlK
BlogDocsLog inGet started
Tessl Logo

talk-farley-vibe-coding-best-we-can-do

Summarizes, explains, and answers questions about Dave Farley's talk 'Vibe Coding — Is this really the best we can do?', including key arguments, frameworks, and recommendations. Provides verbatim-cited explanations of Farley's three properties of programming languages, audits AI-coding setups against his three-problems framework, drafts BDD-style executable specifications, and outlines his continuous-delivery approach to AI-driven development. Use when the user asks about Dave Farley's talk on vibe coding, agentic programming, AI-generated tests, BDD-style executable specifications, problem-specific DSLs, why natural language is insufficient as a programming language, the three properties of programming languages (formal grammar / unambiguous intent / deterministic execution), the three problems AI programming creates (precise specification, verification, incrementalism), fifth-generation programming, AI as compiler, or applying Farley's continuous-delivery approach to AI coding agents.

70

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

77%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 well-structured, concise, and offers clear sequenced workflows with strong validation checkpoints across several use cases. Its main weakness is progressive disclosure: it points to outline.md/transcript.md/quote.md files that are not present in any bundle directory, making the references unverifiable.

Suggestions

Include the referenced reference files (outline.md, transcript.md, quote.md) in a references/ directory so the cross-references resolve to real bundle files.

Consolidate the repeated citation/quote-safety guidance (it currently appears in both the grounding rules and the default workflow) to reduce redundancy.

State explicitly what to do when outline.md lacks a matching section anchor (e.g. fall back to scanning transcript.md), to close a small actionability gap.

DimensionReasoningScore

Conciseness

Mostly lean and assumes Claude's competence without restating basic concepts, but the description field and several enumerated grounding rules repeat similar citation guidance multiple times, leaving minor tightening opportunities.

4 / 5

Actionability

Concrete, executable guidance predominates (read outline.md → locate section → read transcript.md range → answer with cited safe excerpts), though it is instruction-driven rather than code-driven and some steps rely on named file lookups without stating the fallback if a file lacks the expected anchor.

4 / 5

Workflow Clarity

The default workflow plus per-use-case sections are clearly sequenced with explicit verification checkpoints ('If the claim isn't in transcript.md, say...'), feedback loops ('If the framework genuinely doesn't fit, say so'), and per-dimension verdict requirements, satisfying the validation-bar guidance.

5 / 5

Progressive Disclosure

The body references outline.md, transcript.md, and quote.md as detailed material, but no references/, scripts/, or assets/ bundle directories exist, so the referenced files appear to be missing from the bundle and the cross-references cannot be confirmed as one-level-deep real files.

2 / 5

Total

15

/

20

Passed

Description

95%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 exceptionally specific, listing concrete actions and a comprehensive set of natural trigger phrases tied to a single identifiable talk. It cleanly answers both 'what' and 'when' with low conflict risk, though its length approaches verbosity.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('Summarizes, explains, and answers questions', 'audits AI-coding setups against his three-problems framework', 'drafts BDD-style executable specifications', 'outlines his continuous-delivery approach') with comprehensive coverage of the skill's capabilities.

5 / 5

Completeness

Explicitly answers both 'what' (the summary/explain/audit/draft actions) and 'when' via a 'Use when the user asks about...' clause enumerating concrete trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive natural keywords users would say ('vibe coding', 'agentic programming', 'AI-generated tests', 'BDD-style executable specifications', 'problem-specific DSLs', 'continuous-delivery approach') including synonyms and the talk title verbatim.

5 / 5

Distinctiveness Conflict Risk

A clear niche tied to a single specific talk and named frameworks (three properties, three problems), with triggers unlikely to fire for unrelated skills; minimal conflict risk.

5 / 5

Total

20

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository
jscraik/Agent-Skills
Reviewed

Table of Contents

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.