claude

sdlc-core

The six-stage AI-native SDLC loop (Plan, Design, Build, Test, Deploy, Maintain) from a Confluence PRD to a verified draft PR: PRD gate with product-owner feedback and intent.md, human-approved spec, accepted plan, test-first build, fail-to-pass and plan evidence, PR review loop, control bands and triage, learning. Hooks enforce the gates.

0 on GitHubby Engineering Platformupdated 21 days agodevelopment
/

Plugin sdlc-core from marketplace ai-sdlc in amarkum/prd-to-pr-rule-skills. Commands read from the repo's manifest — 19 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (10)

  • context-gather

    Builds the evidence-backed context pack for a story before the spec is written - affected code, data schemas, contracts, prior bugs, incidents and team knowledge - with a citation for every claim. Use after a PRD is READY, or when asked "what does this change touch", "what could break", or "what do

  • learn

    Turns what a story taught the team (review comments, CI failures, rework, PRD clarifications, incidents) into reviewed knowledge entries, CLAUDE.md lines, rules, checks or evals. Use after review feedback or a merge, after an incident, or for "capture lessons" or "run a retro".

  • plan

    Turns an approved spec into plan.md (files that change, test-first tasks, risks, proof per AC, locked tests) in plan mode, so nothing is edited until the plan is accepted. Use after spec approval, or for "plan the work" or "split this into tasks".

  • pr-babysit

    Drives an open AI-assisted PR to green: fixes failing checks and in-scope review comments with evidence re-run, answers every thread, and stops when only code-owner approval is left. Never merges. Use after pr-open, or for "address the review" or "fix CI".

  • pr-open

    The Deploy stage: opens a draft PR carrying what reviewers need (Jira and PRD links, intent, spec, plan, AC coverage, evidence at the SHA, risks, rollback) and links it back to Jira. Use when verify has passed and the user asks to open the PR.

  • prd-review

    The Plan-stage gate: checks a PRD (Confluence page or Markdown) for completeness, testability and ambiguity, returns READY or NEEDS_CLARIFICATION with questions for the product owner, and on READY writes intent.md. Use to review, validate or score a PRD, and as the first stage of prd-to-pr.

  • prd-to-pr

    Runs one Jira story or epic through the AI-native loop (Plan, Design, Build, Test, Deploy, Maintain) from a Confluence PRD to a draft PR with evidence, stopping at every human gate and resuming from specs/<KEY>/. Use for "take this PRD to a PR" or "start PAY-123".

  • spec-write

    The Design stage: writes spec.md from an accepted intent, the PRD and the context pack (traceable requirements, Given/When/Then ACs with IDs, design, repo impact and sequencing, test plan, rollout, flagged policy concerns) as a draft for human approval. Use after context-gather, or for "write the sp

  • triage

    The Maintain stage: diagnoses a production or process signal (a breached control band, Sentry issue, alert, bug ticket or incident thread) read-only and writes it as intent.md; small fixes may go to a draft PR through the normal gates, and fixed incidents get a regression eval. Use for "investigate

  • verify

    The Test stage: runs the repo's checks, maps each acceptance criterion to passing tests, proves new tests fail without the change, checks the diff against plan.md and its locked tests, and records it all at the commit SHA; fixes failures in at most two rounds. Use before a PR, after build, or for "i

Subagents (5)

  • builder

    The team's specialised builder for Airflow data pipelines, micro-frontends and the frontend shell. Implements one task of an accepted plan.md at a time, test-first, following the repo manifest and the stack skill, and commits green. Use for the build stage of prd-to-pr.

  • pr-reviewer

    Adversarial, read-only code reviewer that runs the REVIEW.md passes (bugs, security, compliance with spec.md and plan.md) on a branch diff and returns Important findings and capped nits with file:line and fixes. Use for self-review before pr-open and in CI review jobs.

  • prd-analyst

    Read-only requirements analyst. Reviews a PRD for contradictions, untestable criteria, missing roles and states, data feasibility and conflicts with production history, and returns severity-rated findings with product-owner questions. Use for the semantic half of prd-review.

  • qa-verifier

    Independent, read-only verifier with a fresh context. Re-checks evidence integrity, judges whether the tests actually prove each acceptance criterion (no tautologies, no mocking of the unit under test, unhappy paths covered), and can run the app to check behaviour against plan.md. Use after verify p

  • repo-investigator

    Read-only codebase and history investigator. Maps entry points, similar features, contracts, data shapes, tests and git history relevant to a story, returning facts with file:line citations. Use during context-gather, or whenever you need to know what a change touches before designing it.

Hooks (5)

  • SessionStart

    python3

  • PreToolUse

    python3

  • PreToolUse

    python3

  • PreToolUse

    python3

  • Stop

    python3

View source on GitHub

More from amarkum/prd-to-pr-rule-skills

All 6 plugins →

Atlassian and Sentry MCP plus schema and production-history lookup skills.

2 skills · 2 MCP servers

0

Airflow DAG conventions and test patterns.

1 skill

0

React/TypeScript conventions and Playwright evidence.

1 skill

0

stack-mfe

claude

Module Federation contracts and isolation (builds on stack-frontend).

1 skill

0

Everything in the standard: sdlc-core, the connectors and all stack packs. The one plugin to enable for the whole organization.

0

Similar plugins

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

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

Maintain sdlc-core?

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 plugins6 plugins