Igor Ganapolsky
Community@igorganapolsky · Florida, USA
Production AI reliability for voice agents and agent workflows. Evidence-backed diagnostics, failure containment, and acceptance-tested pilots.
Agent Skills by Igor Ganapolsky
Showing 156 vetted skills indexed across 3 GitHub repositories.
IgorGanapolskyspeckit-checklist
Generate requirements-quality checklists for spec-kit feature specifications.
IgorGanapolskygsd-new-project
Initializes new projects through questioning, research, requirements, and roadmap generation.
IgorGanapolskycompound-lite
Captures engineering learnings into docs/solutions notes after non-trivial work.
IgorGanapolskyusing-git-worktrees
Creates isolated git worktree workspaces with native tool detection and baseline test verification.
IgorGanapolskyprinciple-boundary-discipline
Concentrates validation and error handling at system boundaries while keeping business logic in pure functions.
IgorGanapolskyvalue-center-lite
Applies VSM five-question value-center analysis to planning and release decisions.
IgorGanapolskyprinciple-model-the-domain
Encodes domain logic in data structures instead of scattered conditionals.
IgorGanapolskymaintainability-gap-lite
Detects AI copy-paste code sprawl and enforces refactoring over duplication.
IgorGanapolskytest-driven-development
Enforces red-green-refactor test-driven development workflow before writing production code.
IgorGanapolskyspec-governance-lite
Classifies tasks by difficulty and enforces staged spec-driven development with attributable drift review.
IgorGanapolskygsd-phase
Adds, inserts, removes, and edits phases in ROADMAP.md through routed workflows.
IgorGanapolskysystematic-debugging
Diagnose bugs through a four-phase root cause investigation process before proposing fixes.
IgorGanapolskyprinciple-fix-root-causes
Guides debugging by tracing symptoms to root causes instead of applying symptom-level fixes.
IgorGanapolskyprinciple-encode-lessons-in-structure
Converts recurring corrections into lint rules, metadata flags, runtime checks, or automation scripts.
IgorGanapolskygpt6-astra-harness-lite
Applies Astra harness rules for computer-use automation, session notes, and action confirmation gates.
IgorGanapolskyspeckit-bug-assess
Assess bug reports against the codebase and write structured assessment files.
IgorGanapolskyspeckit-constitution
Creates or updates the project constitution file from resolved Spec Kit templates and user inputs.
IgorGanapolskyusing-superpowers
Enforces skill discovery and invocation before any AI response or action.
IgorGanapolskydispatching-parallel-agents
Dispatch concurrent subagents to fix independent test failures across separate problem domains.
IgorGanapolskyspeckit-implement
Executes all tasks defined in tasks.md following the spec-kit implementation plan.
IgorGanapolskyworkflow-economics-lite
Validates AI workflow pilots against ownership, baseline, and economic metric requirements.
IgorGanapolskyagentzip-memory-lite
Caps agent-sandbox fanout and compresses redundant pages using a memory gate script.
IgorGanapolskyprinciple-guard-the-context-window
Routes large outputs to subagents and keeps summaries in the main context window.
IgorGanapolskyprinciple-subtract-before-you-add
Guides removal of dead code and redundant logic before adding new features.
Frequently Asked Questions About Igor Ganapolsky
FAQPage SchemaWhat tasks can I accomplish with Igor Ganapolsky's skills?▼
You can enforce agent reliability via ThumbGate feedback-to-prevention rules, run Codex adversarial reviews, execute Spec Kit spec-driven development phases, verify revenue claims against Stripe and Plausible telemetry, and govern live trading with position sizing, anomaly detection, and circuit breakers.
Who should use these skills?▼
Engineers running Claude Code or Cursor agent sessions who need failure containment, developers practicing TDD and spec-driven development with review checkpoints, and quantitative traders needing pre-trade verification, sentiment aggregation, and portfolio risk monitoring with budget-aware execution.
How do I install and run these skills?▼
Skills install into Claude Code or MCP-compatible agent sessions as markdown frontmatter definitions. Spec Kit skills require a .specify/ project structure; trading skills need Alpaca or yfinance credentials; gate scripts like speckit_gate.py and superpowers_gate.py run before completion claims.
Are these skills free and open source?▼
Yes, the skills are published publicly on GitHub across 130 repositories. Many skills explicitly enforce fail-closed cost caps, denying paid vendor spend such as Semrush, Dataiku, Neo4j, and Copilot Pro in favor of local, free-tier alternatives under a stated $20 monthly cap.
What prerequisites do these skills require?▼
Prerequisites vary: Spec Kit skills need the .specify/ directory structure, analytics skills need Plausible, PostHog, or GA4 environment variables, trading skills need market data credentials, and several skills depend on companion gate scripts, git worktrees, or Docker for local CI via ACT.