Altivum's eval -> plan -> execute -> deploy feature-development pipeline as one refinable, versioned workflow. Provides a /ship orchestrator, five analysis lenses (eval, improve, improve-ui, improve-x2, enhance), a Workflow-powered /recon multi-lens audit, and a /refine continuous quality-gated loop (parallel build, live functional verification, auto-PR + auto-merge on green). Includes read-only codebase-analyzer/ui-auditor/frontier-researcher/functional-verifier and deploy-validator/security-reviewer subagents and a blocking production-deploy gate.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Deploy the implemented change safely — feasibility check, reviewable/reversible mechanisms, migrations, and live verification.
UX-friction lens — traces real user flows and proposes up to 3 code-grounded fixes for the friction users actually hit, ranked by impact.
Empirical health & architecture audit — structure, correctness, security, maintainability — with prioritized, evidence-backed fixes.
Execute a previously generated implementation plan with precision, TDD, and verification.
Interface-craft lens — audits visual design quality and modern CSS, proposes exactly 3 UI upgrades with before/after code.
Read-only deep analysis of ONE focus area of a codebase (a layer like frontend/backend/data, or a cross-cutting concern such as UX-flow friction). Returns structured findings — strengths, gaps, file:line evidence, candidate improvements with rough impact/effort. Does not rank or edit. Invoke one per focus area, in parallel.
Use after a deployment to verify it actually worked end-to-end — checks infra/app status, live endpoints, deployed artifacts, and migrated data. Read-only; reports evidence, never mutates infrastructure.
Read-only live web research on ONE angle (category leaders, signature UX/interaction patterns, emerging platform capabilities, or a field-specific question) for a given project positioning. Returns concrete cited findings (name + URL + technique + why-now) grounded to the repo. Invoke one per angle, in parallel.
Functional gate for a code change. Launches the app and OPERATES the changed functionality end-to-end — web changes via Claude-in-Chrome, non-web changes via their real interface (API/CLI/function) — then reports whether a real user could actually use it. The critical must-pass of the refine quality gate. Runs processes and drives a browser, but never edits source code.
Use during/after the execute phase to review code changes for security vulnerabilities — auth/authz, injection, input validation, secrets, IAM least-privilege, data exposure, and unsafe external calls. Read-only; reports findings, does not modify code.
Executes bash commands
Hook triggers when Bash tool is used
Uses power tools
Uses Bash, Write, or Edit tools
Public Claude Code plugin marketplace from Altivum Inc. —
open workflow tooling for the community. Marketplace name: altivum.
claude plugin marketplace add AltivumInc-Admin/claude-plugins-public
claude plugin install altivum-feature-dev-pipeline@altivum
Update later by re-running claude plugin install (or claude plugin update) after a new release.
| Plugin | Version | What it does |
|---|---|---|
altivum-feature-dev-pipeline | 0.2.0 | A full feature workflow: /ship orchestrates eval → plan → execute → deploy with approval gates between phases. Includes deploy-validator & security-reviewer subagents and a blocking pre-deploy gate. |
/altivum-feature-dev-pipeline:ship <goal> — runs the whole pipeline, pausing for your approval between phases.:eval, :plan, :execute, :deploy.security-reviewer subagent reviews diffs for auth/IDOR, injection, secrets, IAM least-privilege, and more.deploy-validator subagent verifies a deploy end-to-end (status, live endpoints, migrated data).See the plugin's README for full details.
claude-plugins-public/
├── .claude-plugin/marketplace.json
└── plugins/
└── altivum-feature-dev-pipeline/
├── .claude-plugin/plugin.json
└── commands/ agents/ hooks/ README.md
Apache-2.0 © Altivum Inc.
This is the public distribution repo. Plugins are developed in a private repo and released here once vetted.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub altivuminc-admin/claude-plugins-marketplace --plugin altivum-feature-dev-pipelineIntent-anchored refinement loop. The inventor authors intent (what the software is for); sensors measure the gap between code and intent (allowed to find nothing); the inventor picks the quarter-turn; the actuator builds it under a hard quality gate (live functional operation, security, adversarial review) and auto-PRs/auto-merges on green; a journal remembers every cycle. Signal never creates work — only the inventor does. Supersedes altivum-feature-dev-pipeline.
Altivum's eval -> plan -> execute -> deploy feature-development pipeline as one refinable, versioned workflow. Provides a /ship orchestrator, five analysis lenses (eval, improve, improve-ui, improve-x2, enhance), a Workflow-powered /recon multi-lens audit, and a /refine continuous quality-gated loop (parallel build, live functional verification, auto-PR + auto-merge on green). Includes read-only codebase-analyzer/ui-auditor/frontier-researcher/functional-verifier and deploy-validator/security-reviewer subagents and a blocking production-deploy gate.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.