Skip to content
View shmindmaster's full-sized avatar
2 followers · 0 following

Achievements

Achievement: Pull Sharkx4Achievement: Pair Extraordinairex4Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pull Sharkx4Achievement: Pair Extraordinairex4Achievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Block or report shmindmaster

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
shmindmaster/README.md

Sarosh Hussain

I build trustworthy enterprise AI systems and the developer infrastructure that makes agent work verifiable.

CTO & Partner at Pendoah · Former CIO, Perry Homes · Director-level AI/cloud transformation at Accenture, PwC, and Omnicom/Credera.

Thesis: production AI needs architecture and control planes—agent safety, exact evidence, governance, and delivery discipline—not demos.

Open-source products (released)

CrewScore · crewscore.ai

AI-agent prompt governance linter — finds missing written safeguards in system prompts.

  • 23 published controls across 8 dimensions (human approval, injection defense, cost limits, stop conditions, auditability, and more)
  • Offline, deterministic, no API key / no LLM
  • Python CLI · GitHub Action · GitHub Marketplace · browser-local checker
  • Current release: v0.6.11 (PyPI + GitHub Release)
  • Public corpus study: 356 prompts scored offline; among 83 production-labeled prompts, median coverage 10/100 (coverage ≠ quality; methodology is published)

pip install crewscore · Action: shmindmaster/crewscore@v2

GitPin · site

Agent-delivery assurance gate + local evidence MCP — makes agent-authored claims point to exact committed source before merge.

  • Read-only, multi-repo, private/offline Git roots
  • 12 pin.* MCP tools: candidates → evidence packs → independent verification
  • Exact path, line, full commit SHA, content hash; dirty worktrees excluded from committed evidence
  • PR evidence gate (base-trusted policy, changed-path coverage)
  • Current release: v0.6.3 on npm, GitHub Releases, GitHub Pages, and official MCP Registry (io.github.shmindmaster/gitpin)
  • Formerly RepoContext 0.3.x (superseded product lineage)

npx -y gitpin@0.6.3 · MCP identity: io.github.shmindmaster/gitpin

What I do in production

Enterprise AI architecture and delivery—agentic platforms, RAG/LLMOps, regulated workflows—and the unglamorous work of making systems survive real estates, budgets, and audits. Client identities under NDA.

Open source is where I publish the control-plane pieces: written-control governance (CrewScore) and commit-pinned agent evidence (GitPin).

Houston, TX · saroshhussain.com · LinkedIn · crewscore.ai

Popular repositories Loading

  1. crewscore crewscore Public

    Find the safety rules your AI agent prompt forgot. Offline linter — 23 controls, CLI + Action + browser. No API key.

    Python 1

  2. gitpin gitpin Public

    Require exact PR evidence before agent-authored changes merge. Local evidence MCP + base-trusted gate.

    TypeScript

  3. shmindmaster shmindmaster Public

    Profile README

  4. awesome-mcp-servers awesome-mcp-servers Public

    Forked from punkpeye/awesome-mcp-servers

    A collection of MCP servers.