corticalstack
Community@corticalstack · Geneva, CH
corticalstack provides explicit-invocation engineering skills for git workflows, phased implementation planning, code review, document processing, and LLM evaluation.
Agent Skills by corticalstack
Showing 31 vetted skills indexed across 2 GitHub repositories.
corticalstackinit
Bootstraps a project with AGENTS.md, tracker configuration, and cross-vendor review profiles.
corticalstackcontent-research-writer
Guide topic clarification, outlining, drafting, and iterative refinement with citations.
corticalstackcommit
Creates staged git commits with planned messages after user confirmation.
corticalstackresume-handoff
Resumes work from handoff documents by validating current codebase state against recorded context.
corticalstackcontext-degradation
Detect and diagnose context degradation patterns in long prompts for LLM agents.
corticalstackscientific-brainstorming
Generate research ideas and structured exploration pathways from an under-specified scientific problem.
corticalstackresearch-requirements
Collect constraints, technology preferences, and success criteria from issue or text briefs.
corticalstackresearch-codebase
Documents the existing codebase via parallel read-only sub-agents and writes findings to flow/research/.
corticalstackfastapi-expert
Generate production-grade FastAPI async REST API guidance with Pydantic v2 and JWT/OAuth2.
corticalstackexploratory-data-analysis
Detect scientific file types and generate markdown EDA reports.
corticalstackcreate-handoff
Writes a structured handoff document under flow/handoffs/ to transfer work between sessions.
corticalstackcreate-plan
Creates phased, test-driven implementation plans under flow/plans/ from issues or task descriptions.
corticalstackhandle-pr-feedback
Fetch PR review feedback, implement requested changes, and push updates for re-review.
corticalstackxlsx
Create and update Excel workbooks with formulas, pandas analysis, and LibreOffice recalculation.
corticalstackconsciousness-council
Conduct structured multi-perspective deliberation to resolve decision dilemmas and creative trade-offs.
corticalstackExtract text and tables from PDFs and fill fillable or non-fillable forms.
corticalstackiterate-plan
Revises existing implementation plans in flow/plans/ based on user feedback and codebase research.
corticalstackcross-review
Runs an advisory cross-vendor code review of the current branch diff against main.
corticalstackcontext-fundamentals
Guide AI agent context design and debugging under token limits.
corticalstackpython-pro
Creates Python 3.12+ async code with strict typing and test coverage.
corticalstackvalidate-plan
Validate an implementation against its plan by running verification commands and reporting deviations.
corticalstackpptx
Convert, create, and edit PowerPoint presentations via OOXML-level modifications.
corticalstackterraform-engineer
Implement Azure Terraform modules with Azure Blob remote state and validation.
corticalstackchangelog-generator
Convert git commit history into categorized, user-friendly changelog sections.
Frequently Asked Questions About corticalstack
FAQPage SchemaWhat tasks can I accomplish using corticalstack's skills?▼
You can stage and create git commits, convert GitHub issues into phased implementation plans, handle PR feedback, run cross-vendor code reviews, process PDF/DOCX/XLSX/PPTX files, perform exploratory data analysis, and score LLM outputs with debiased evaluation rubrics.
Who are corticalstack's skills designed for?▼
They target software engineers and technical leads managing git-based delivery cycles, plus data analysts and researchers needing document processing, EDA reporting, scientific brainstorming, and structured LLM output evaluation with bias mitigation.
How do corticalstack's skills run in practice?▼
Most skills are explicit-only commands invoked via slash commands like /commit, /create-plan, or /cross-review, with disable-model-invocation set to true. They never fire autonomously, and planning skills write artifacts under flow/plans/, flow/handoffs/, and flow/research/.
What prerequisites or dependencies do these skills require?▼
Skills operate on a git working tree and GitHub pull requests. The init skill bootstraps projects by creating AGENTS.md and .claude/tracker.json, while cross-review uses OAuth-backed keyless backends such as Azure Foundry, GitHub Models, or Copilot CLI.
Do corticalstack's skills support cloud infrastructure work?▼
Yes. The terraform-engineer skill implements Azure Terraform modules with Azure Blob remote state and validation, and fastapi-expert generates production-grade async REST services with Pydantic v2 and JWT/OAuth2 authentication patterns.