Skip to content
SkillModel-led2,021 tokens in use

Content Eval

Turn podcast transcripts, meeting notes, sales calls and competitor scans into 20 to 30 content ideas, score them with a seven-expert panel and get a ranked menu.

3,862 stars on the repository MIT Licenseericosiu/ai-marketing-skills · content-eval
npx skills add ericosiu/ai-marketing-skills -s content-eval
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Content Eval costs

31tokens idle
2,021tokens in use
3,196tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter8/15
  • Structure23/25
  • Examples9/15
  • Output10/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity88
  • Maintenance44
  • Community56
  • Maturity28
  • Momentum40

Content Eval is a weekly content ideation and scoring pipeline. It gathers raw material from recent podcast episodes, meeting notes, sales call insights and any topics you mention, runs a competitive scan of YouTube channels when you have not supplied one, and generates 20 to 30 ideas across long-form video, Shorts and X or LinkedIn articles.

Each idea must serve a messaging pillar, then passes through a seven-expert panel covering hooks, algorithm fit, brand fit, buyer relevance, differentiation, short-form adaptation and debate. Ideas averaging 85 or higher pass, are ranked and are placed in a four-week production schedule. The rest go on a kill list with the reason.

What You Get

  • A results document with executive summary, competitive gaps, ranked passing ideas, schedule, kill list and raw material summary
  • A scoring table per idea with seven expert scores, a top insight and the biggest weakness
  • A four-week production schedule that moves from low effort and high impact toward strategic anchors
  • A kill list with each failed idea’s average score and the dimension that sank it
  • An optional team channel summary with the top five ideas and the week one plan

Common Use Cases

  • Plan a week of YouTube long-form, Shorts and articles from what you recorded and heard
  • Decide what to film by scoring a list of ideas you already have
  • Find topics no competitor covers, using outlier videos and format patterns
  • Turn recurring client questions from calls and meetings into content

What Makes It Unique

It joins idea generation and scoring in one pass and forces every idea to earn a pillar. The panel has hard scoring integrity rules: generic ideas that anyone could make score under 70 on differentiation, and ideas without personal receipts or data score under 75 on brand fit. The 85 threshold exists so only the strongest ideas survive, and failures are listed with reasons.

Under The Hood

It is a prompt-led skill with four reference files and no scripts of its own. The agent reads local folders of podcast transcripts and meeting notes, checks your call recording data, and may call a YouTube competitive analysis tool for the scan. It then creates a results document and can optionally post a summary to your team channel. It reads whatever sources you point it at, so keep sensitive recordings out of folders you do not want summarized.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers