~ / rigs / jthiruveedula / agentkit

jthiruveedula/agentkit

Portable, vendor-neutral agent foundation for Claude Code, Copilot, Cursor, and Antigravity — one command restores your skills/agents/commands on any machine.

↗ GitHub ★ 0 MIT updated 1d ago personal setup Claude CodeCodex
share on X
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
~7.1k tokens
Heavy · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details
Quality 0/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit jthiruveedula/agentkit/agents ./rig-agentkit/agents
$ npx degit jthiruveedula/agentkit/skills ./rig-agentkit/skills

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ npx degit jthiruveedula/agentkit/skills/agentkit-audit .claude/skills/agentkit-audit
$ npx degit jthiruveedula/agentkit/skills/chart-design .claude/skills/chart-design
$ npx degit jthiruveedula/agentkit/skills/context-budget .claude/skills/context-budget
$ npx degit jthiruveedula/agentkit/skills/context-compressor .claude/skills/context-compressor
$ npx degit jthiruveedula/agentkit/skills/daily-standup .claude/skills/daily-standup
$ npx degit jthiruveedula/agentkit/skills/data-architect .claude/skills/data-architect
$ npx degit jthiruveedula/agentkit/skills/data-eng-router .claude/skills/data-eng-router
$ npx degit jthiruveedula/agentkit/skills/data-quality-standards .claude/skills/data-quality-standards
$ npx degit jthiruveedula/agentkit/skills/data-storytelling .claude/skills/data-storytelling
$ npx degit jthiruveedula/agentkit/skills/debug-loop .claude/skills/debug-loop
$ npx degit jthiruveedula/agentkit/skills/doc-standards .claude/skills/doc-standards
$ npx degit jthiruveedula/agentkit/skills/ext-archify .claude/skills/ext-archify
$ npx degit jthiruveedula/agentkit/skills/ext-aws .claude/skills/ext-aws
$ npx degit jthiruveedula/agentkit/skills/ext-azure .claude/skills/ext-azure
$ npx degit jthiruveedula/agentkit/skills/ext-cli-ops .claude/skills/ext-cli-ops
$ npx degit jthiruveedula/agentkit/skills/ext-codegraph .claude/skills/ext-codegraph
$ npx degit jthiruveedula/agentkit/skills/ext-databricks .claude/skills/ext-databricks
$ npx degit jthiruveedula/agentkit/skills/ext-diagrams .claude/skills/ext-diagrams
$ npx degit jthiruveedula/agentkit/skills/ext-gap-discovery .claude/skills/ext-gap-discovery
$ npx degit jthiruveedula/agentkit/skills/ext-gcp .claude/skills/ext-gcp
$ npx degit jthiruveedula/agentkit/skills/ext-hallmark .claude/skills/ext-hallmark
$ npx degit jthiruveedula/agentkit/skills/ext-meta-muse .claude/skills/ext-meta-muse
$ npx degit jthiruveedula/agentkit/skills/ext-rea .claude/skills/ext-rea
$ npx degit jthiruveedula/agentkit/skills/ext-security-audit .claude/skills/ext-security-audit
$ npx degit jthiruveedula/agentkit/skills/ext-seo .claude/skills/ext-seo
$ npx degit jthiruveedula/agentkit/skills/ext-slides .claude/skills/ext-slides
$ npx degit jthiruveedula/agentkit/skills/ext-systems-design .claude/skills/ext-systems-design
$ npx degit jthiruveedula/agentkit/skills/ext-ui-ux .claude/skills/ext-ui-ux
$ npx degit jthiruveedula/agentkit/skills/frontend-motion-3d .claude/skills/frontend-motion-3d
$ npx degit jthiruveedula/agentkit/skills/genai-engineering .claude/skills/genai-engineering
$ npx degit jthiruveedula/agentkit/skills/humanizer .claude/skills/humanizer
$ npx degit jthiruveedula/agentkit/skills/mcp-security-review .claude/skills/mcp-security-review
$ npx degit jthiruveedula/agentkit/skills/meeting-to-actions .claude/skills/meeting-to-actions
$ npx degit jthiruveedula/agentkit/skills/memory-sync .claude/skills/memory-sync
$ npx degit jthiruveedula/agentkit/skills/memory .claude/skills/memory
$ npx degit jthiruveedula/agentkit/skills/ml-lifecycle .claude/skills/ml-lifecycle
$ npx degit jthiruveedula/agentkit/skills/orchestrate .claude/skills/orchestrate
$ npx degit jthiruveedula/agentkit/skills/pipeline-scaffold .claude/skills/pipeline-scaffold
$ npx degit jthiruveedula/agentkit/skills/pr-review .claude/skills/pr-review
$ npx degit jthiruveedula/agentkit/skills/prompt-enhancer .claude/skills/prompt-enhancer
$ curl -fsSL --create-dirs -o .claude/agents/data-platform-architect.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/data-platform-architect.md
$ curl -fsSL --create-dirs -o .claude/agents/data-quality-engineer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/data-quality-engineer.md
$ curl -fsSL --create-dirs -o .claude/agents/doc-writer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/doc-writer.md
$ curl -fsSL --create-dirs -o .claude/agents/frontend-engineer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/frontend-engineer.md
$ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.<redacted>.md
$ curl -fsSL --create-dirs -o .claude/agents/ml-engineer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/ml-engineer.md
$ curl -fsSL --create-dirs -o .claude/agents/orchestrator.md https://raw.githubusercontent.<redacted>.md
$ curl -fsSL --create-dirs -o .claude/agents/pipeline-engineer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/pipeline-engineer.md
$ curl -fsSL --create-dirs -o .claude/agents/researcher.md https://raw.githubusercontent.<redacted>.md
$ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/test-writer.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/test-writer.md
$ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/jthiruveedula/agentkit/main/agents/verifier.md

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(./.env.*)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(git push --force:*)",
      "Read(**/*.pem)",
      "Read(.env)",
      "Read(~/.aws/**)",
      "Read(**/*.key)",
      "Read(**/.env.*)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Bash(rm -rf *)",
      "Bash(sudo *)",
      "Bash(sudo:*)",
      "Bash(rm -rf /)",
      "Bash(git push -f:*)",
      "Read(.env.*)",
      "Bash(git reset --hard*)",
      "Bash(rm -rf /*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(gh pr merge *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(gh pr merge:*)",
      "Bash(gh pr create *)",
      "Bash(git reset --hard:*)",
      "Bash(npm publish:*)",
      "Bash(git rebase *)"
    ]
  }
}

Skills (46)

Subagents (12)

data-platform-architect
model: sonnet
Decides data platform architecture (storage, compute, orchestration, batch vs streaming) and records ADRs; writes no pipeline code. Use when choosing a warehouse, orchestrator, or streaming approach.
data-quality-engineer
model: sonnet
Adds schema, freshness, volume, idempotency, and PII checks to a data pipeline for the risks flagged. Use when a pipeline is about to ship.
doc-writer
model: sonnet
Updates README/CHANGELOG/reference docs to match a completed change; never edits source. Use when a completed change alters user-facing behavior.
frontend-engineer
model: sonnet
Builds UI, motion (CSS, Motion, GSAP), and 3D (three.js/R3F) within a performance and a11y budget. Use for animation, scroll effects, WebGL, or interactive UI work.
implementer
model: sonnet
Makes code changes from a spec or researcher findings, scoped to the named files. Use when the change is already scoped with acceptance criteria.
ml-engineer
model: sonnet
Implements ML and GenAI work (training/eval, RAG, LLM calls, eval harnesses) against a named metric. Use for model, RAG, agent, or LLM-cost tasks.
orchestrator
model: inherit
Lead agent: writes a validated plan, fans tasks to roster subagents in parallel, verifies, synthesizes. Use for multi-specialty goals, not a single focused change.
pipeline-engineer
model: sonnet
Implements a dbt model, Airflow DAG, Dagster asset, or PySpark job from an already-decided architecture. Use when the ADR exists and transform logic is needed.
researcher
model: sonnet
Read-only investigation: locates code, gathers facts, compares options; cites file:line or URL. Use when you need facts before any edit.
reviewer
model: sonnet
Reviews a diff/PR for correctness bugs and scope creep; read-only, reports findings only. Use when a diff is ready for review, after implementation and before merge.
test-writer
model: sonnet
Writes the smallest test that fails if the logic under test breaks. Use when `implementer` flags coverage gaps.
verifier
model: sonnet
Runs external checks only (tests, linters, type checkers, build) and reports pass/fail with decisive output; never edits files. Use to prove a change works.

Similar rigs

copied ✓