claude

ai-native-sdlc

Ten advisory skills covering the six stages of the AI-native SDLC: intent, spec, plan, CLAUDE.md, feedback loop, evals, PR review, approval gates, and closing the loop, plus a project audit advisor.

0 on GitHubby Shawn Sandyv1.0.0MITupdated 2 months agoengineering
/

Plugin ai-native-sdlc from marketplace ai-native-engine in shawn-sandy/ai-native-engine. Commands read from the repo's manifest — 1 month ago.

What it installs

Skills (10)

  • agent-evals

    Build and maintain a continuous eval suite that regression-tests the agent's configuration (CLAUDE.md, skills, hooks, model, prompts) in CI (Stage 4 of the AI-native SDLC). Use when asked to set up evals, gate changes to .claude/**, turn an incident into a regression test, or check whether a model o

  • ai-native-sdlc-advisor

    Advises on adopting the AI-native SDLC. Audits a repo's agentic setup (CLAUDE.md, skills, hooks, plan/intent artifacts, evals, review) and recommends next plays; also nudges at stage boundaries during normal work.

  • approval-gates

    Express the human approvals a change process requires as Claude Code hooks that allow, ask, or block (Stage 5 of the AI-native SDLC), and place build-time guardrails behind advisory skills. Use when asked to gate deploys, protect paths or test files, set up hooks, or harden settings for a team or re

  • claude-md-maintainer

    Create, trim, and keep CLAUDE.md current as the repo's institutional knowledge (Stage 3 of the AI-native SDLC). Use when asked to set up or improve CLAUDE.md, when a mistake has been made twice in a session, or when a review flags that CLAUDE.md is outdated.

  • close-the-loop

    Design the autonomous maintenance loop where a deterministic detector watches a production or process metric and invokes Claude by tier — log, diagnose read-only, or propose a PR/runbook — with findings written back as intent.md (Stage 6 of the AI-native SDLC). Use when asked to set up monitoring ag

  • feedback-loop

    Give every session a way to verify its own work before a human sees it (Stage 4 of the AI-native SDLC) — one-command build/test/lint, a quantifiable done target, failing-test-first for bug fixes, screenshot loops for UI, and protection for the tests themselves. Use when setting up verification, fixi

  • intent-capture

    Turn an idea, ticket, or incident into a committed intent.md (Stage 1 of the AI-native SDLC). Use when someone describes a problem or wants to "write this up", "capture the intent", or start a feature/fix with no spec yet.

  • plan-from-spec

    Start implementation in plan mode from an approved spec.md and produce a committed plan.md naming files, order of work, risks, and proof (Stage 3 of the AI-native SDLC). Use when asked to "plan this", "start the build", or when a spec.md/intent.md is handed over for implementation.

  • pr-review-policy

    Write or apply a REVIEW.md review policy — bugs, security, and compliance passes against spec.md/plan.md, severity rules, nit caps — and run the PR fix loop (Stage 5 of the AI-native SDLC). Use when asked to review a PR, draft review instructions, address review comments, or babysit a PR to green.

  • spec-from-intent

    Produce a requirements-and-design spec.md from an accepted intent.md, applying the organization's brand/security/compliance/UX skills and flagging every concern (Stage 2 of the AI-native SDLC). Use when asked to "spec this out", "write requirements", or when an intent.md is attached or referenced.

View source on GitHub

Similar plugins

mattpocock-skills

claude

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

agents-md

claude

AGENTS.md as project instructions, by one option. Under claude-md-or-agents-md a project with no instruction files of its own gets its AGENTS.md files instead: a prompt.context hook asks $.fs.ancestors for every AGENTS.md and .claude/AGENTS.md from the filesystem root down to the working directory and hands them to the engine as project instruction files, which it renders, announces and withholds from agents exactly as it does CLAUDE.md, skipping any file the engine already loaded by import or link; a tool.call hook on Read attaches the ones under the project root as the engine attaches a nested CLAUDE.md. Under claude-md-and-agents-md every AGENTS.md is loaded beside CLAUDE.md. Under managed-only the project's and the person's instruction files are dropped and the organization's managed ones kept. Under claude-md the engine's walk stands alone. claude-md-or-agents-md is the default.

anthropics/claude-code 155,476

agent-skills

claudecodex

Production-grade engineering skills covering every phase of software development: spec, plan, build, verify, review, and ship.

25 skills · 4 agents

addyosmani/agent-skills 116,782

agentic-qe-fleet

claude

Agentic QE (external: maintained in proffesor-for-testing/agentic-qe, pinned to v3.15.17): a quality-engineering fleet with 11 QE agents, 9 skills, 9 commands, the agentic-qe MCP server pinned to the plugin version, and the aqe-mod guard that refuses destructive operations on AQE learning data

9 skills · 9 commands · 11 agents · 1 MCP server

ruvnet/ruflo 81,589

reverse-engineering

claudecodex

Binary reverse engineering, malware analysis, firmware security, and software protection research for authorized security research, CTF competitions, and defensive security

4 skills · 3 agents

wshobson/agents 42,461

Maintain ai-native-sdlc?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted