j1ngg/tech-marketing-framework / .claude/skills/blog
blog
VerifiedGenerates SEO and AEO optimized blog posts from source content. Supports feature announcements, product launches, thought leadership, cookbooks/tutorials, report conversions, and news analysis. Outputs complete posts with TLDR, FAQ, schema markup, and internal linking recommendations.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 25.5 KB |
| autoresearch-blog/SKILL.md.baseline | other | 25.1 KB |
| autoresearch-blog/blog-optimized-v1.md | markdown | 25.5 KB |
| autoresearch-blog/changelog.md | markdown | 1.6 KB |
| autoresearch-blog/dashboard.html | other | 6.2 KB |
| autoresearch-blog/results.json | config | 944 B |
| autoresearch-blog/results.tsv | other | 256 B |
autoresearch-blog/changelog.md
View on GitHub# Autoresearch Changelog: Blog Skill
## Configuration
- **Target skill:** blog/SKILL.md
- **Test inputs:** 5 (feature announcement, product launch, thought leadership, cookbook/tutorial, news analysis)
- **Evals (v2, tightened):** 5 (reference value, "I didn't know that", naming the pain, no filler, formatting discipline)
- **Runs per experiment:** 5
- **Max score:** 25
---
## Experiment 0 — baseline (tightened evals)
**Score:** 24/25 (96%)
**Change:** None. Original skill, no modifications.
**Result:** 4 of 5 outputs passed all evals. Feature announcement failed the "no filler" eval: the three use cases (AI agents, CI/CD, service accounts) all made the same point (scoped ephemeral credentials applied to different contexts) without showing distinct capabilities.
**Failing outputs:** Feature announcement use cases section.
---
## Experiment 1 — keep
**Score:** 25/25 (100%)
**Change:** Added anti-repetition instruction to the Feature Announcement use cases section: "Each use case must highlight a DIFFERENT capability or behavior of the feature. If two use cases make the same point, merge them or replace the duplicate."
**Reasoning:** The baseline failure was the model repeating "ephemeral credentials" across all use cases. An explicit instruction to differentiate should force distinct angles.
**Result:** Use cases now show genuinely different capabilities: (1) contextual source environment verification for CI/CD, (2) purpose attestation at runtime for third-party tokens. The post explicitly distinguishes them. No filler eval now passes.
**Failing outputs:** None.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |