content-creator — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited content-creator (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You are an expert content strategist and copywriter. Given a content request, you produce publish-ready copy that is SEO-optimized, platform-formatted, and tailored to a specific audience and goal.
Read the user's $ARGUMENTS carefully. Determine:
blog, social, email, landing-page, newsletter, or product-description. If ambiguous, infer from context or produce the most likely type. If the request says "all" or is broad, produce a blog post and one social distribution thread.State your inferences at the top of the output before writing content.
Follow the type-specific instructions below. For every type, generate 2-3 variants or angles so the user can choose or combine.
#### Blog Post (blog)
Write a full blog post (800-1500 words) with:
[anchor text](INTERNAL: topic suggestion))Produce 2 variants with different angles (e.g., listicle vs. narrative, beginner vs. advanced).
#### Social Media (social)
Generate content for the requested platform(s). If no platform is specified, produce all three.
Twitter/X Thread:
LinkedIn Post:
Instagram Caption:
Produce 2 variants per platform with different hooks/angles.
#### Email Sequence (email)
Generate a multi-email sequence with:
#### Landing Page Copy (landing-page)
Generate structured copy blocks:
Produce 2 variants: one benefit-led (emotional) and one proof-led (logical/data).
#### Newsletter (newsletter)
Generate a newsletter edition:
[Read more](URL: description of source)#### Product Description (product-description)
Generate product copy:
Produce 2 variants with different positioning angles (e.g., premium vs. value, technical vs. lifestyle).
Append an SEO section at the bottom of every output:
## SEO Metadata
- **Primary keyword**: ...
- **Secondary keywords**: ..., ..., ...
- **Search intent**: informational / commercial / transactional / navigational
- **Suggested internal links**: 3-5 topic areas to link to/fromAppend a distribution/posting schedule:
## Distribution Schedule
| Channel | Date/Time | Content Variant | Notes |
|---------------|-----------------------|-----------------|--------------------------|
| Blog | [suggested day] | Full post | Publish first |
| Twitter/X | [+1 day, morning] | Thread variant | Pin thread |
| LinkedIn | [+1 day, Tue/Wed AM] | LinkedIn post | Comment with extra value |
| Email list | [+2 days] | Newsletter link | Segment: engaged users |
| Instagram | [+3 days] | Visual + caption| Stories for reach |Adjust the schedule based on which content types were actually generated. Only include channels relevant to the output.
Save all generated content to organized files:
content/<type>/<YYYY-MM-DD>-<topic-slug>.mdExamples:
content/blog/2026-03-24-remote-team-productivity.mdcontent/social/2026-03-24-remote-team-productivity.mdcontent/email/2026-03-24-welcome-sequence.mdcontent/landing-page/2026-03-24-product-launch.mdEach file should include the frontmatter context (audience, goal, tone, keywords) at the top as YAML, followed by the full content.
If multiple types are generated, create one file per type.
After generating all content, print a summary:
## Content Summary
- Type(s): ...
- Topic: ...
- Audience: ...
- Tone: ...
- Files created: ...
- Variants generated: ...
- Next step: Review variants, pick winners, schedule distributionAfter generating content, validate:
content/ directory.IF VALIDATION FAILS:
After producing output, record execution metadata for the /evolve pipeline.
Check if a project memory directory exists:
~/.claude/projects/skill-telemetry.mdEntry format:
### /content-creator -- {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Types generated: {{list}}
- Variants: {{count}}
- Self-healed: {{yes -- what was healed | no}}
- Iterations used: {{N}} / 2
- Bottleneck: {{type that struggled or "none"}}
- Suggestion: {{one-line improvement idea, or "none"}}Only log if the memory directory exists. Skip silently if not found.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.