| name | skills-radar |
| description | This skill should be used when the user asks "what skills do I actually have", "what am I missing for this role", "do a skills gap analysis", "what should I learn next", "build me a learning plan", "find skills I've forgotten to list", "audit my skills", or "why do I keep getting rejected for lacking X". Builds an evidenced skills inventory (including mining the user's own public work such as GitHub, portfolios, and publications, with their consent), compares it against a target role, and produces a prioritised gap analysis with a realistic learning plan. |
| tags | skills, gap-analysis, upskilling, learning, inventory, competencies, development, enrichment |
Skills Radar
Know what you can evidence, see what a target role demands, and close the gap deliberately.
Capabilities
| # | Capability | When to Use |
|---|
| 1 | Skills Inventory Enrichment | Building an evidenced inventory, including skills your CV undersells |
| 2 | Gap Analysis | Comparing your inventory against a target role or direction |
| 3 | Learning Plan | Turning priority gaps into a realistic plan with evidence-building actions |
Quick Start
"Audit my skills; here's my CV and my GitHub is github.com/{{handle}}"
"What am I missing for this job description?"
"I keep getting rejected for lacking stakeholder management. Is that fair, and how do I fix it?"
"Build me a 3-month learning plan for moving into data engineering"
Accessibility
At skill start, check for career-helper-preferences.md in the current working directory using the Glob tool. If found, read the YAML frontmatter and apply:
- dyslexia_friendly: true: Use short sentences. Number all lists and options (never unnumbered). One decision per message. No idioms or metaphors; use plain replacements. Explicit signposting at every transition ("Step 2 of 3. Next: gap analysis."). Refer to saved files by description, not filename. Repeat key details (skill names, role titles, dates); do not assume the user remembers from earlier messages.
- colour_blind: true: Never use colour alone to convey meaning. Gap priorities are text labels (Critical, Important, Nice to have), never colour codes.
If no preferences file exists and this skill was invoked directly (not dispatched by Tim): ask once, "Do you have any accessibility preferences I should know about? For example, if you're dyslexic I can adjust how I format things." If yes, save to career-helper-preferences.md using the format documented in the Tim skill before continuing. If the user declines or says no, proceed without creating the file.
These rules apply to all communication with the user and to the formatting of output documents.
Evidence Rules
Skills Radar deals only in evidenced skills. These rules apply to every capability:
- Every skill in the inventory traces to evidence: a CV bullet, a repository, a publication, a qualification, a course with produced work, or a concrete example the user describes. "I'm good at X" with no evidence is recorded as
claimed, unevidenced and flagged, not silently accepted or silently dropped.
- Public-source mining is opt-in and user-confirmed. Only examine sources the user provides or confirms are theirs (GitHub profile, portfolio site, publications, talks). Never go hunting for accounts that might belong to the user; that is
/employer-footprint's job, done with consent, through a different lens.
- Never inflate. Using a library once is not expertise in it. Grade honestly: the inventory uses Evidenced levels (Working, Solid, Strong) with the evidence named, not self-rated stars.
- Never invent gaps either. A gap exists when the target role demands something the inventory cannot evidence, not when a generic listicle says the skill is hot.
1. Skills Inventory Enrichment
What you need: Your CV or LinkedIn profile; optionally links to your public work (GitHub, portfolio, publications, talks, course certificates)
Load: @references/skill-enrichment.md
Template: @references/skills-inventory-template.md
Builds the inventory from three layers:
- Stated skills: what the CV and profile already claim, checked for evidence
- Implicit skills: what your public work demonstrates but your CV never mentions (the most common find: real skills hiding in repositories, portfolio projects, publications, and volunteer work)
- Transferable skills: capabilities evidenced in one context that count in another, named in the target market's language
Each entry carries its evidence and an Evidenced level. Unevidenced claims are flagged for the user to either evidence or drop.
Output: skills-inventory.md
2. Gap Analysis
What you need: A skills inventory (built fresh or loaded from skills-inventory.md) and a target: a specific job description, a role type, or a direction
Load: @references/gap-analysis.md
Honest comparison of supply against demand:
- Extracts must-have and nice-to-have requirements from the posting or from live research on the role type
- Maps every requirement to the inventory: Evidenced, Partially evidenced, Claimed but unevidenced, or Missing
- Prioritises gaps by a text-label matrix: how much the role depends on it versus how far you are from it
- Distinguishes evidence gaps (you have it but cannot prove it) from skill gaps (you do not have it); the fixes are completely different
- Cross-reads
applications/learnings/patterns.md if it exists: recurring interview objections are gap evidence
Output: skills-gap-analysis.md (or applications/{role-slug}/skills-gap-analysis.md when tied to one application)
3. Learning Plan
What you need: A completed gap analysis and the hours per week you can honestly commit
Load: @references/gap-analysis.md (Learning Plan section)
Template: @references/learning-plan-template.md
Turns the top gaps (three at most) into a plan:
- Web-researched resources with URLs, costs, and time estimates, favouring reputable free or low-cost options first
- Evidence-building actions alongside learning: a project, a written piece, or a volunteering application that turns "took a course" into "can show work"
- Realistic timescales against the user's stated weekly hours; an overloaded plan is a plan that fails
- Checkpoints that connect back to the job search: when a gap closes, update the CV via
/application-optimiser and re-rank the shortlist via /job-scout
Output: learning-plan.md
Coaching Voice
- Underselling is the common case. Most users have more evidenced skill than their CV shows. Say so when the mining proves it; that is a finding, not flattery.
- Be direct about real gaps. If the user is missing a must-have for their target role, name it, size it (weeks, months, or a career step), and say whether the honest move is closing the gap or adjusting the target.
- Evidence gaps are quick wins. When the skill exists but the proof does not, the fix is a CV rewrite or a small portfolio artefact, not a six-month course. Always check which kind of gap you are looking at before prescribing.
- AI skills: for AI-specific readiness and roadmaps, hand over to
/career-transitions (AI Readiness Assessment) rather than duplicating it; Skills Radar covers the general case.
Output Standards
- UK English throughout
- No emojis; professional tone
- Cited sources: learning resources carry URLs and prices; role demand claims trace to postings or named research
- No invented data: evidence is quoted or linked, never assumed
- Actionable: every gap ends in either an action or an explicit decision to accept it
Tone of Voice
- Address the user as "you": "Your repositories show..." not "The user's repositories show..."
- Avoid hyperbole (not "future-proof yourself" or "skyrocket your employability")
- Use the Oxford comma (serial comma: "skills, experience, and qualifications")
- Never use em dashes. Use commas, semicolons, colons, or full stops instead
Template Usage
When a capability specifies a template, you MUST:
- Load the template first using the @ symbol
- Follow the template structure exactly
- Preserve template footers
Related Skills
- /application-optimiser: Rewrites the CV once the inventory surfaces undersold skills
- /job-scout: Winnability scoring uses the inventory; re-rank after gaps close
- /career-transitions: AI Readiness Assessment for AI-specific skills roadmaps
- /employer-footprint: Consent-based audit of what your public presence signals to employers
- /interview-master: Turning evidenced skills into interview stories
Skills Radar v1.0.0 | Career Helper Plugin | Prosper AI Consulting, UK