Travis Sparks
Community@sparkst · Sammamish, WA
I build AI solutions and coach software development and product teams how to build scalable systems with AI
Agent Skills by Travis Sparks
Showing 57 vetted skills indexed across 1 GitHub repositories.
sparkstQCOE
Guides structured Correction of Errors analysis with 5-Whys root cause investigation and COE document generation.
sparkstDiscovery Call Assistant
Captures structured notes, follow-up questions, and pain points during live discovery calls.
sparkstInterview Synthesis
Transforms discovery call notes into quick wins, customized guides, ranked solutions, and pilot recommendations.
sparkstQDOC - Progressive Documentation
Generate and update project documentation following a three-tier progressive docs pattern.
sparkstQUX - UX Test Scenarios
Generate UX test scenarios for UI components covering happy paths, edge cases, errors, and accessibility.
sparkstQIDEA - Research and Ideation
Research topics across multiple sources and generate options matrices with recommendations.
sparkstQGIT - Git Release Management
Run quality gates, generate Conventional Commit messages, and push git changes.
sparkstIndustry Signal Scout
Searches web sources with Tavily and Brave to filter noise and find authoritative evidence.
sparkstResearch Planning
Creates structured research plans with sub-questions, query strategies, claim budgets, and source tier rules.
sparkstOptions Matrix Builder
Synthesizes specialist disagreements into structured options matrices with pros, cons, risk, and cost analysis.
sparkstFact Checking Protocol
Validates research claims against tiered sources and checks date alignment with a Python script.
sparkstSource Quality Policy
Evaluates source credibility using a four-tier classification system with independence and recency checks.
sparkstWeb Research Executor
Orchestrates parallel Tavily and Brave web searches with deduplication and caching.
sparkstqralph
Orchestrates a deterministic multi-agent pipeline from ideation through deployment and smoke testing.
sparkstux-design-pipeline
Orchestrates wireframing, multi-critic UX review, component implementation, and browser testing for UI screens.
sparkstCOS Request Intake
Parses Chief of Staff requests and generates staffing plans with specialist assignments.
sparkstBuy vs Build Decision Matrix
Evaluates build-vs-buy decisions across seven weighted dimensions with a 3-year TCO calculator.
sparkstAmazon PR-FAQ Generator
Generates Amazon-style PR-FAQ documents with press release, FAQs, and premortem.
sparkstPMF Validation Plan
Designs product-market fit validation plans using JTBD interviews, WTP surveys, and retention cohorts.
sparkstTenets Framework
Creates non-negotiable team tenets with rationale, trade-offs, and pass/fail examples.
sparkstSecurity Review for Third-Party Skills
Reviews third-party Claude Skills for security risks using tiered checklists.
sparkstStory Point Estimator
Estimates story points for tasks using Fibonacci and fine-grained scales with complexity adjustments.
sparkstPrincipal Engineer Code Reviewer
Reviews code changes for correctness, security, and standards compliance with structured JSON findings.
sparkstVideo Transcript
Extract audio from video files and transcribe to markdown with speaker diarization using FFmpeg and whisperX.
Frequently Asked Questions About Travis Sparks
FAQPage SchemaWhat tasks can I accomplish with sparkst's skill collection?▼
You can run multi-agent code reviews (QCHECK, QFULL-REVIEW, review-pr), execute TDD cycles (QCODET/QCODE), manage git releases with Conventional Commits (QGIT), conduct root-cause COE analyses, perform tiered source fact-checking, and generate branded content including LinkedIn carousels, infographics, and hero images.
Who are these skills designed for?▼
Software engineers, principal-level reviewers, product managers, and technical content creators. Engineering skills target Supabase Edge Functions and TDD teams, while COS skills (PR-FAQ, buy-vs-build, PMF validation) serve product leads, and QWRITE/QPPT serve newsletter and LinkedIn publishers.
How do I install and run these skills?▼
Skills install into a Claude skills directory (e.g., ~/.claude/skills) and activate via named triggers such as QCHECK, QWRITE, or QPPT. Each skill declares its claude_tools (Read, Grep, Bash, etc.), agents, and references in frontmatter; pipelines like QRALPH orchestrate phases end-to-end.
What prerequisites and dependencies are required?▼
Several skills require Python 3.8+ (fact-check, web-exec, buy-vs-build). Video transcription needs FFmpeg and whisperX; hero-image generation uses Playwright and Gemini; web research uses Tavily and brave-search with network access. Most review and planning skills have no external dependencies.
Are sparkst's skills open source and free to use?▼
The manifest specifies no licensing or pricing terms; the skills are distributed as Claude skill folders with versioned frontmatter (mostly 1.0.0) under the public sparkst GitHub profile. Check the individual repositories for license files before commercial redistribution.