| name | blog-idea-generator |
| description | Use when generating and prioritising blog topics, editorial angles, or a topic-ideas file from client, audience, search, and website context; use blog-writer only after an idea is selected. |
| metadata | {"portable":true,"compatible_with":["claude-code","codex"]} |
Blog Idea Generator
Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com, +256 784 464178.
Use When
- Use this skill when the user wants blog topics, editorial angles, or content ideas.
- Load it before article drafting when ideation and prioritization are still needed.
Do Not Use When
- The user already chose an article and needs full drafting.
- The request is about proposals or unrelated to content publishing.
Required Inputs
- The topic area, audience, and publishing objective.
- Any company, sector, geography, or SEO context that should shape the ideas.
Domain Method
- Gather the business context, audience, and publishing purpose.
- Assess what information is already available and what ideation gaps remain.
- Use the workflow below to generate, refine, and prioritize candidate ideas.
- Return idea summaries that can be handed off cleanly to
blog-writer/.
Quality Standards
- Produce differentiated ideas with clear audience and angle fit.
- Balance SEO usefulness with authority-building and originality.
- Preserve compatibility with existing repository workflows and file paths.
Domain Risks
- Do not return generic topic lists with no angle or audience fit.
- Do not skip summary or prioritization when the user needs a working shortlist.
- Do not treat ideation as if it were the full article-writing workflow.
Existing Deliverables
- A prioritized list of blog ideas and short article-ready summaries.
Do Not Use When
- Do not use this skill when a neighbouring specialist named in the description owns the primary decision; route there first and use this skill only as a supporting layer.
- Do not use it to invent buyer facts, evidence, legal positions, statutory rules, delivery capacity, or current external claims.
Inputs
| Artefact | Source/provider | Required? | Missing-input behaviour |
|---|
| website or brand context, audience, goals, existing content, search evidence, and editorial constraints | Buyer, ToR, approved project record, accountable owner, or verified evidence source | Yes | Stop the affected decision, name the missing source, and return only a qualified outline or assumption register. |
Outputs
| Artefact | Consumer | Acceptance condition |
|---|
| prioritised blog idea set with narrative briefs | Editor, subject-matter expert, and blog writer | Claims trace to supplied evidence; assumptions, owners, exclusions, decisions, and observable acceptance tests are explicit. |
Evidence Produced
| Evidence | Consumer | Acceptance condition |
|---|
| Decision and source trace for the output | Reviewer and release owner | Each load-bearing choice identifies its source, rationale, accountable owner, and any unassessed check. |
Capability Contract
Default to read-only for analysis, critique, discovery, and review. Minimum capability is access to the supplied artefacts and permission to inspect or calculate relevant evidence. Edit only the requested working copy. Do not publish, send, spend, change production systems, certify compliance, make a statutory claim, or approve a commercial concession without explicit authority.
Degraded Mode
If required files, interviews, finance doctrine, search evidence, calculation tools, network access, or specialist review are unavailable, return the narrowest useful qualified result. Mark unavailable checks as not assessed, separate facts from assumptions, and state what is needed to resume. An unassessed gate is never a pass.
Decision Rules
| Choice | Action | Failure/risk avoided |
|---|
| Generate, merge, reject, or prioritise an idea | Score audience relevance, evidence availability, differentiation, search intent, and editorial fit. | A high-volume list of generic topics. |
| Evidence conflicts or authority is absent | Stop the affected recommendation, record the conflict, and seek the named owner’s decision. | Invented facts or unauthorised commitments. |
| Evidence and approval are complete | Proceed within scope and retain the decision trace. | Irreproducible approval or scope drift. |
Workflow
- Confirm the consumer, decision, neighbouring-skill route, authority, and required inputs; stop if the primary route or accountable owner is unknown.
- Inspect supplied evidence and record facts, assumptions, conflicts, and unavailable checks; stop on a load-bearing contradiction.
- Apply the domain method and decision rules, preserving the repository’s proposal voice and specialist constraints.
- Draft the contracted output with observable acceptance conditions and an evidence trace.
- Review alignment with scope, work plan, team, pricing, risk, and dependent proposal sections; recover by revising the affected choice and rerunning the check.
- Run the critical-analysis and anti-slop gates; block release on unsupported claims, failed safety or finance gates, or an F slop grade.
Quality Standards
- Every load-bearing claim is verified or explicitly qualified, and the output distinguishes fact, assumption, recommendation, and commitment.
- Scope, method, work plan, staffing, pricing, risks, dependencies, and acceptance conditions remain mutually consistent.
- The output preserves domain constraints, names accountable owners, covers failure paths, and blocks unsupported or unauthorised promises.
- British English and the repository’s East African professional tone are used unless the buyer requires another standard.
- The critical-analysis and anti-slop gates pass before release, with no unassessed check represented as passed.
Anti-Patterns
- Inventing a buyer fact or proof point. Fix: cite the supplied source or mark the statement as an assumption requiring confirmation.
- Treating an unavailable check as passed. Fix: mark it not assessed and name the evidence needed to resume.
- Copying a neighbouring skill’s method without routing. Fix: use the specialist for the primary decision and retain only the supporting layer here.
- Writing acceptance as “satisfactory” or “appropriate”. Fix: state an observable measure, evidence record, and decision owner.
- Adding premium or technical claims without delivery capacity. Fix: reconcile the claim with named people, time, budget, dependencies, and authority.
Worked Example
For an East African accounting SaaS, reject “Benefits of Technology” and prioritise a sourced article on reconciling mobile-money settlements, with a named audience, evidence plan, and distinct angle.
References
- Local
references/ files for ideation frameworks and content formats.
- blog-writer when an idea moves into drafting.
Generate 15-25 targeted blog post ideas, each presented as a 200-word hybrid summary with narrative brief + structured specs. The system adapts its ideation methods to the specific client and available information.
Read ideation-frameworks for the full 20-method library and selection logic.
Read content-formats for 20 content formats with structural templates.
Step 1: Gather Context
Read client docs (mandatory)
Read every available file to build a complete picture:
docs/en/company-profile.md (and all enabled language versions)
docs/en/services.md — service offerings, target customers
docs/en/pages.md — existing website pages and content
docs/sector-brief.md — industry context (if present)
docs/style-brief.md — brand voice and tone
- topic-ideas — existing topics (avoid duplicates)
src/pages/en/blog/ — existing articles (avoid overlap)
- All other
docs/en/ files — testimonials, FAQ, portfolio, about-story
Extract and note:
- What the business does (core services, products)
- Who they serve (audience segments, industries, company sizes)
- Where they operate (geographic focus, markets)
- What makes them different (competitive advantage, methodology)
- What expertise the author has (experience, credentials, stories)
- What problems customers face (pain points, challenges)
- What content already exists (published articles, covered topics)
Guided interview (3-5 questions)
After reading docs, ask targeted questions to fill gaps. Ask one at a time. Skip questions already answered by docs.
Core questions (ask what's missing):
- Audience specifics — "Who is your ideal reader? (Job title, company size, industry, location)"
- Top pain points — "What are the top 3 problems your customers face that your business solves?"
- Content goals — "What should readers DO after reading? (Contact you, book a demo, understand a concept?)"
- Competitor landscape — "Name 2-3 competitors. What topics do they cover?"
- Unique knowledge — "What do you know that competitors don't? What's your unfair advantage?"
- Customer questions — "What questions do customers ask most before buying?"
- Content gaps — "Topics you've wanted to write about but haven't?"
- Context/audience — any additional context the user provides (specific themes, campaigns, seasonal needs)
If the user provides additional context (audience details, campaign goals, seasonal focus), incorporate it into the assessment.
Step 2: Assess Available Information
Score each dimension to determine which ideation methods will work best:
| Dimension | Rich (3) | Moderate (2) | Sparse (1) |
|---|
| Client docs | Detailed company-profile, services, testimonials, stories | Basic company-profile and services | Minimal — just a business name and description |
| Competitor visibility | Named competitors with active blogs | Competitors named but blogs unknown | No competitor info |
| Audience specificity | Named segments with pain points | General audience description | Vague ("businesses") |
| Industry dynamism | Active news cycle, regulations, trends | Moderate change rate | Stable/static industry |
| Existing content | 5+ published articles to spin off | 1-4 articles | No existing content |
| Customer interaction | Direct customer questions available | Some FAQ data | No customer feedback |
Step 3: Select Ideation Methods
Based on the assessment, select 5-7 methods from the 20-method library. Always include Methods 1 and 2 as foundation.
Selection Matrix
| Method | Best When | Min Score |
|---|
| 1. Category Drilldown | Always | — (always include) |
| 2. Buyer Awareness Stages | Always | — (always include) |
| 3. Pain Point Mining | Client docs ≥ 2 or customer interaction ≥ 2 | — |
| 4. Competitor Gap Analysis | Competitor visibility ≥ 2 | Competitor 2+ |
| 5. Customer Question Mapping | Customer interaction ≥ 2 | Customer 2+ |
| 6. They Ask, You Answer | Customer interaction = 3 | Customer 3 |
| 7. Amazon/Review Mining | Product-based business | Client docs 2+ |
| 8. Spin-Off Posts | Existing content ≥ 2 | Content 2+ |
| 9. Media Mashup | Brand voice is informal/creative | Client docs 2+ |
| 10. Highlight Good/Bad | Industry has notable examples | Industry 2+ |
| 11. How-To/Tutorial Mining | Product/service has teachable processes | Client docs 2+ |
| 12. Success/Failure Stories | Client has real project stories | Client docs 3 |
| 13. Holiday/Event Mapping | Content calendar needs seasonal hooks | Any |
| 14. Newsjacking/Trends | Industry dynamism = 3 | Industry 3 |
| 15. Use Any Object | Need creative/lateral ideas | Any (creative fallback) |
| 16. Curated Roundups | Industry has notable resources | Industry 2+ |
| 17. Prediction Posts | Industry dynamism ≥ 2 | Industry 2+ |
| 18. Jargon/Glossary | Technical niche with newcomer audience | Audience 2+ |
| 19. Contrarian/Negative | Audience is sophisticated | Audience 3 |
| 20. Topic-Category Matrix | Need high volume quickly | Any (volume fallback) |
Announce: "Based on available information, I'm using methods: [list]. Here's why: [brief rationale]."
Step 4: Generate Ideas
Run selected methods sequentially. Aim for 25-35 raw ideas, then filter to the best 15-25.
For each method, consult ideation-frameworks for detailed instructions and examples.
Quality Filters
Remove any idea that fails:
| Filter | Test |
|---|
| High-value goal | Does this help the reader make/save money, reduce risk, save time, or gain advantage? |
| Unique angle | Does this require knowledge that isn't commonly available? |
| So-what test | Would the target reader care enough to click? |
| Longevity | Will this still be relevant in 12 months? |
| No overlap | Not already published or in existing topic-ideas.md? |
| Searchable | Would someone type this into a search engine? |
Tier Classification
| Tier | Purpose | Target Count |
|---|
| Tier 1: SEO drivers | Attract organic traffic via long-tail keywords | 6-8 ideas |
| Tier 2: Authority builders | Establish expertise with deep guides and analysis | 5-7 ideas |
| Tier 3: Thought leadership | Build brand with opinions, predictions, stories | 4-5 ideas |
Step 5: Create 200-Word Hybrid Summaries
For each approved idea, produce a summary in this exact format:
### [Number]. [Working Title]
[3-4 sentence narrative brief: What this article is about, who it serves,
why it matters now, and the unique angle that makes it worth reading. This
paragraph should make someone want to write — and read — this article. It
captures the creative direction and emotional tone.]
- **Audience:** [specific reader segment — job title, industry, company size]
- **Buyer Stage:** [Awareness / Consideration / Decision]
- **Format:** [How-to / Case study / List / Opinion / Guide / Story / Comparison / Interview / Roundup / FAQ]
- **Angle:** [the specific twist that differentiates from competitors — 1 sentence]
- **Key Points:**
1. [what the article must cover — specific enough to outline from]
2. [second key point]
3. [third key point]
4. [fourth key point — optional]
5. [fifth key point — optional]
- **CTA Goal:** [what action the reader should take after reading]
- **SEO Keywords:** [primary keyword], [secondary keyword]
- **Tier:** [1: SEO driver / 2: Authority builder / 3: Thought leadership]
- **Est. Words:** [1,500-2,500]
Summary Quality Rules
- The narrative must read like a creative brief — not a dry description
- Key points must be specific enough to outline section headings from
- Keywords must be realistic long-tail phrases someone would search
- The angle must be genuinely different from what a Google search would surface
- Every title must pass the 4 U's test: Useful, Unique, Urgent, and Ultra-specific; score 3+ on at least three dimensions.
Step 6: Present and Refine
Present to the User
Show ideas grouped by tier with full summaries. After presenting, ask:
- Which ideas excite you most?
- Any ideas to remove or modify?
- Any topics you expected but don't see?
- Any specific campaigns or seasonal needs to address?
Refine based on feedback. The user's input overrides the assessment.
Step 7: Save Output
Save the final approved list to topic-ideas:
# Blog Topic Ideas — [Client Name]
Generated: YYYY-MM-DD
Methods used: [list of methods applied]
Target audience: [summary]
Content categories: [list]
## Tier 1: SEO Drivers
### 1. [Title]
[Full 200-word hybrid summary as above]
## Tier 2: Authority Builders
...
## Tier 3: Thought Leadership
...
## Content Calendar Suggestion
| Month | Article 1 (Tier) | Article 2 (Tier) |
|-------|-------------------|-------------------|
| Month 1 | [title] (T1) | [title] (T2) |
...
If the file already exists, merge new ideas — don't overwrite existing topics. Mark previously written topics as [PUBLISHED].
Quality Checklist
Before finalising:
After writing, verify line count is under 500: wc -l blog-idea-generator/SKILL.md