Sanjan B M
Community@sanjanb · Banglore
AI & ML Engineer
Agent Skills by Sanjan B M
Showing 104 vetted skills indexed across 1 GitHub repositories.
sanjanbpair-agent
Generates a setup key to pair a remote AI agent with your browser session.
sanjanbbenchmark
Detects web performance regressions by benchmarking page load metrics against stored baselines.
sanjanbdesign-html
Generates production-quality HTML and CSS from approved design mockups using Pretext text layout.
sanjanbbrowse
Controls a persistent headless Chromium browser for QA testing, screenshots, and page state verification.
sanjanbios-clean
Removes the DebugBridge SPM package and debug wiring from iOS apps.
sanjanbplan-tune
Tune AskUserQuestion preferences and inspect developer behavior profiles across gstack skills.
sanjanbdesign-shotgun
Generates multiple AI design variants and collects structured feedback through a comparison board.
sanjanbplan-design-review
Reviews UI/UX design plans across seven scored passes and fixes gaps interactively.
sanjanbhandoff
Generates a handoff document summarizing the current conversation for a fresh agent session.
sanjanbdesign-consultation
Generates a complete design system and DESIGN.md source of truth for projects.
sanjanbto-tickets
Breaks plans and specs into tracer-bullet tickets with declared blocking edges published to a tracker.
sanjanbautoplan
Runs CEO, design, engineering, and DX plan reviews sequentially with automated decisions.
sanjanblearn
Review, search, prune, and export project learnings stored across sessions.
sanjanbguard
Warns on destructive bash commands and restricts file edits to a chosen directory.
sanjanbresearch
Investigates questions against primary sources and writes cited findings to a Markdown file.
sanjanbprototype
Build throwaway prototypes to answer design questions about logic or UI.
sanjanbfreeze
Restricts Edit and Write operations to a user-specified directory for the session.
sanjanbgit-guardrails-claude-code
Configures Claude Code hooks to block dangerous git commands before execution.
sanjanbtriage
Move GitHub issues and external PRs through a triage state machine with agent-ready briefs.
sanjanbios-qa
Runs vision-driven QA loops against SwiftUI apps on a physical iPhone over USB.
sanjanbcareful
Intercepts destructive bash commands and prompts for confirmation before execution.
sanjanbopen-gstack-browser
Launches an AI-controlled Chromium browser with a live sidebar activity feed.
sanjanbcso
Audits repositories for secrets, supply chain risks, CI/CD flaws, and OWASP vulnerabilities.
sanjanbplan-ceo-review
Reviews product plans with CEO-level scope expansion, reduction, and premise-challenging analysis.
Frequently Asked Questions About Sanjan B M
FAQPage SchemaWhat tasks can I accomplish with Sanjan B M's skill registry?▼
The registry covers end-to-end software delivery: spec writing, plan reviews (CEO, engineering, design, DX), test-driven development, web and iOS QA, security audits, performance benchmarking, deployment pipelines, retrospectives, and documentation generation. Skills also handle web scraping, PDF export, diagram creation, and cross-model benchmarking.
Who is the target audience for these skills?▼
The skills target software engineers, engineering managers, founders, and designers working with AI coding agents such as Claude Code and OpenCode. iOS developers building SwiftUI apps, web developers needing QA and performance regression detection, and teams wanting structured plan reviews and retrospectives benefit most.
How are these skills installed and invoked at runtime?▼
Skills are defined via YAML frontmatter with name, description, triggers, and allowed-tools fields, installed under a skills directory such as ~/.claude/skills/gstack. They activate through natural-language trigger phrases or slash commands, and some register PreToolUse hooks that run shell checks before Edit, Write, or Bash operations execute.
What safety and security capabilities are included?▼
The careful, freeze, guard, and git-guardrails skills warn before destructive commands like rm -rf or force-push, block dangerous git operations, and restrict edits to specific directories. The cso and security-review skills provide OWASP audits, vulnerability scanning, and PR review with confidence scoring and git blame context.
What prerequisites or dependencies do these skills require?▼
Skills assume a host coding agent environment (Claude Code, OpenCode, or Codex CLI) with Bash, Read, Write, Edit, Glob, and Grep tool access. Specific skills need extras: iOS skills require a SwiftUI project with SPM, gbrain setup needs PGLite or Supabase, and deployment skills detect platforms like Fly.io, Vercel, or GitHub Actions.