New:Introducing Socket Scanning for VS Code Marketplace Extensions.Learn more →
Get Started
s

shahvatsal

npm

1.24K

total weekly downloads

Packages

Current Co-maintainers

Former Co-maintainers

17 packages

archsutra-ai

ArchSutra AI: open-source software architecture audit, design and threat-modeling agent skills for Claude Code, Cursor, Codex, GitHub Copilot and Gemini CLI, plus a keyless, evidence-first CLI.

contextsutra

ContextSutra AI: local-first, keyless CLI that builds verifiable, privacy-aware project context capsules plus 12 portable agent skills for Claude Code, Cursor, Codex, Gemini CLI and Copilot.

contribsutra-ai

ContribSutra AI: offline, evidence-backed open-source contributor onboarding CLI plus 25 portable Agent Skills (Claude Code, Cursor, Codex, Copilot, Gemini CLI). Teaches contributors; protects maintainers from PR spam. No API keys, no network.

deploysutra-ai

DeploySutra AI: offline, read-only deployment risk and reversibility analysis for Kubernetes manifests and OpenTofu/Terraform plans. Source-linked diffs, 19 stable policy rules, blast-radius map, a rollback-evidence gate that can never say READY, MD/HTML/

docsutra-ai

Offline documentation auditor: verifies links, anchors, commands, accessibility and leaked secrets in Markdown, and ships 16 portable AI agent skills. No API key, no network, nothing executed.

edgesutra-ai

EdgeSutra AI: evidence-first edge AI compatibility, quantization-planning and benchmark auditing CLI plus 20 portable agent skills for Claude Code, Cursor, Codex, Copilot and Gemini CLI. Keyless, offline, honest about what is NOT measured.

evalsutra

EvalSutra AI: evidence-first, keyless, offline agent evaluation CLI plus 18 portable agent skills for Claude Code, Cursor, Codex, Gemini CLI and Copilot.

incidentsutra

IncidentSutra AI: offline, read-only incident evidence CLI. Builds auditable UTC timelines and competing hypotheses from logs, metrics and deployments, with MD/HTML/JSON reports, an automated audit and 20 portable agent skills for Claude Code, Cursor, Cod

lineagesutra-ai

LineageSutra AI: offline, evidence-first SQL data lineage and schema-change impact CLI with column-level lineage, dbt manifest import, an MCP server, audit reports (MD/HTML/JSON) and 25 portable Agent Skills for Claude Code, Cursor, Codex, Copilot and Gem

localsutra-ai

LocalSutra AI: offline-first local AI readiness CLI. Hardware profile, model-fit estimates, Ollama/llama.cpp loopback diagnostics, honest benchmarks and an independent audit report (MD, HTML, JSON), plus 25 portable Agent Skills. No API keys, no cloud, ze

migrasutra

Evidence-first migration intelligence CLI and 20 portable agent skills. Offline, deterministic, read-only audits of npm, Python, Docker and SQL migrations with MD/HTML/JSON reports.

pramanasutra

PramanaSutra AI: open-source AI QA, test automation and release-readiness agent skills for Claude Code, Cursor, Codex, GitHub Copilot and Gemini CLI, plus a keyless, offline, evidence-first CLI.

protocolsutra

Offline-first MCP and A2A interoperability lab: validate snapshots, audit tool poisoning, diff contracts, replay traces, gate releases. Zero dependencies.

rachnasutra

RachnaSutra AI: open-source UI/UX design, accessibility and responsive audit agent skills for Claude Code, Cursor, Codex, Copilot and Gemini CLI, plus an evidence-first, keyless CLI.

@shahvatsal/pramanasutra

PramanaSutra AI: open-source AI QA, test automation and release-readiness agent skills for Claude Code, Cursor, Codex, GitHub Copilot and Gemini CLI, plus a keyless, offline, evidence-first CLI.

vigilsutra

VigilSutra AI: open-source, offline-first security and privacy auditor CLI plus AI agent skills. Finds secrets, CI/CD, Docker, Kubernetes, IaC, code and AI-agent risks with redacted, evidence-based SARIF/JSON/Markdown reports. No API key, no telemetry.

wattsutra

Honest performance and resource evidence CLI plus 20 portable agent skills. Offline, deterministic benchmarks that never present CPU time as energy; MD/HTML/JSON reports and a self-audit.