Claude Skills by vamseeachanta

End-to-end pipeline for writing ML/AI research papers — from experiment design through analysis, drafting, revision, and submission. Covers NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Integrates automated experiment monitoring, statistical analysis, iterative writing, and citation verification.

Auto-query llm-wiki domains for relevant context before executing domain tasks

Generate clinical trial protocols for medical devices or drugs through

Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to

Run nf-core bioinformatics pipelines (rnaseq, sarek, atacseq) on sequencing

Systematic framework for scientific problem selection, project ideation,

Deep learning for single-cell analysis using scvi-tools: data integration, batch correction, multi-modal analysis, reference mapping, and more.

Performs quality control on single-cell RNA-seq data (.h5ad or .h5 files) using scverse best practices with MAD-based filtering and comprehensive visualizations.

Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.

X/Twitter via xurl CLI: post, search, DM, media, v2 API.

Prevent Codex/Gemini adversarial review dispatch failures caused by relative prompt-file paths and background workdir drift.

Prevent adversarial review dispatch failures caused by relative prompt paths, superseded background sessions, and stale completion notices when launching Codex/Gemini review jobs.

Prevent stale plan/code review prompts from being sent to Codex/Gemini after the underlying artifact changed.

Prevent approval-stage adversarial review churn by keeping plans implementation-focused, schema-complete, and free of stale review-process narration.

Run an approval-stage adversarial cross-review sweep across multiple GitHub issues, drafting any missing canonical plan artifacts before dispatching Codex/Gemini reviews.

Recover adversarial plan reviews on large or sandbox-hostile repos by embedding grounded facts + full plan text directly in a compact prompt and explicitly forbidding reviewer tool use.

Recover adversarial plan reviews when providers keep reading stale remote/main content instead of the revised local artifact.

Add Google-style docstrings to all public functions and classes in a Python package. Uses AST parsing for precise gap detection, priority ranking by coverage ratio, and multi-file patching.

Recover adversarial plan reviews in large or tool-constrained repos by using compact self-contained prompts that forbid tool use and rely only on embedded grounded facts plus the plan text.

Debug Hermes TUI slash commands: Python, gateway, Ink UI.

Handle Gemini adversarial-review runs that emit repeated 429 capacity errors before either recovering with a real verdict or failing unavailable.

Compact live-execution checklist companion for approved GitHub issue work. Use during active implementation as a fast operational route; gh-work-execution remains the canonical source of truth.

Canonical GitHub issue execution route after plan approval — strengthened resource intelligence, TDD-first implementation, targeted validation, adversarial review, delegation controls for Codex agent teams, GitHub progress posting, future-issue capture, and commit/push with closeout discipline.

Audit a repo's live GSD/get-shit-done workflow state against docs, issues, and runtime outputs to find stale issues, automation reliability gaps, parser drift, migration residue, and policy contradictions.

Author in-repo SKILL.md: frontmatter, validator, structure.

Assess a multi-machine, multi-repo AI-enabled workspace for readiness, provider allocation, and issue prioritization.

Dispatch parallel adversarial reviews to Codex and Gemini CLIs for plans or code artifacts. Use when the AI Review Routing Policy requires two- or three-provider review — architecture-heavy, security-affecting, cross-module, or high-stakes changes.

Debug Node.js via --inspect + Chrome DevTools Protocol CLI.

Design self-contained prompts for 3-5 terminals to run overnight without supervision. Ensures zero git contention, provider-optimal allocation, and a clear morning deliverable summary.

Build overnight parallel batches that close stale-open GitHub issues by proving landed work already satisfies the issue, instead of wasting implementation lanes on redoing completed work.

Launch approved GitHub issue implementation in parallel using isolated git worktrees, committed execution-pack prompts, local plan-approved markers, and direct background Codex runs when delegate_task workers are unreliable for real repo writes.

Keep iterative plan-review reruns truthful by syncing prompt files, canonical review artifacts, and the plan's own review-summary narrative after each adversarial wave.

Run a planning-only multi-terminal wave to harden blocked `status:plan-review` issues for fresh adversarial re-review, with zero implementation work and explicit path ownership.

Class-level workflow for plan-review hardening, adversarial review reruns, artifact grounding, approval hygiene, and prompt/path drift recovery.

Plan mode for Hermes — inspect context, write a markdown plan into the active workspace's `.hermes/plans/` directory, and do not execute the work.

Prevent adversarial review dispatch failures caused by sandbox/tmp path mismatches and provider CLI working-directory drift when launching Codex or Gemini with prompt files.

Debug Python: pdb REPL + debugpy remote (DAP).

Prevent stale adversarial re-reviews by forcing Codex/Gemini to review the exact revised local plan artifact instead of stale GitHub issue text or default-branch content.

Pre-commit verification pipeline — static security scan, baseline-aware quality gates, independent reviewer subagent, and auto-fix loop. Use after code changes and before committing, pushing, or opening a PR.

Generate live issue-specific Codex prompts for a single terminal, with repo-aware path contracts and plan-gate safety checks.

Write test suites for SKELETON packages (zero test coverage) to uplift them to DEVELOPMENT maturity. Handles non-importable modules, optional dependencies, separate sub-repos, and Flask/web packages without runtime.

Throwaway experiments to validate an idea before build.

Use when executing implementation plans with independent tasks. Dispatches fresh delegate_task per task with two-stage review (spec compliance then code quality).

Deprecated alias for gh-work-execution.

Run cross-agent memory bridge. Sync learnings between machine agents, update shared knowledge. Phase 5 of #1760.

Import a GSD-2 (.gsd/) project back to GSD v1 (.planning/) format

Join the GSD Discord community

Review and promote backlog items to active milestone

Manage parallel workstreams — list, create, switch, status, progress, complete, and resume

Extract learnings from the current session and convert them to actionable repo content. Phase 4 of #1760.