Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By lucasfcosta
Drive a project goal to completion by enforcing automated quality gates (lint, tests, verification) at every step, using browser automation when needed.
npx claudepluginhub lucasfcosta/backpressured --plugin backpressuredSets up backpressure for a project — detects its check/run commands, writes or edits a BACKPRESSURE.md, verifies Playwright, and installs the MCP servers the loop uses. Runs ONLY when the user explicitly invokes `/backpressured:config`; never auto-trigger it.
Use when a backpressured code-review subagent is reviewing a diff for correctness/logic, simplicity/reuse, or test-quality defects — everything except type design. The default reviewer dimension.
Runs a backpressured automated development loop. Use it ONLY when the user explicitly mentions "backpressured" or "backpressure" — e.g. asking for a "backpressured" loop/run, "do this backpressured", or invoking it by that name — or runs the `/backpressured:goal` command. A bare `/goal`, or any goal-like request that does not mention backpressure, does NOT trigger this loop.
Verifies a feature works by running the real app, not by trusting green tests. Use when a backpressured loop reaches the before-done gate (Phase 3) and needs to confirm new behavior end-to-end — curling live API endpoints and driving a real browser via Playwright — including the cheap unhappy paths, before the task is called done.
Reviews implementation plans during iterative backpressured development. Use when a backpressured loop's Phase 1 reviewer subagent is judging whether a lightweight implementation plan's approach/architecture is sound before any code is written — approving it or sending it back with approach-level concerns.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Self-orchestrating multi-agent development system — 8 specialized AI agents, parallel quality gates, and automated workflows. You say WHAT, the AI decides HOW.
Persona-driven AI development team: orchestrator, team agents, review agents, skills, slash commands, and advisory hooks for Claude Code
HarnessFlow — From idea to shipped product: high-quality engineering workflows for AI agents. Spec-anchored SDD, gated TDD, evidence-based routing, independent reviews, and formal closeout.
Auto-loop execution workflow with quality gates for Claude Code. Automatically decomposes tasks, implements code, runs tests, and iterates through quality gates until completion.
Helder's personal SDLC toolbelt for AI coding agents — from PRD to ship. Bundles the tracer-bullet workflow alongside TDD, code review, audits, and shipping skills.
Requirements-driven development workflow with quality gates for practical feature implementation
Backpressured is a set of skills designed to help agents iterate autonomously for longer. It teaches the agent how to validate its own work with automated checks, adversarial reviews, and even manual testing through cURL or a real browser (via Playwright).
Backpressured runs a four-phase loop:plan, implement, verify, ship and monitor. It gates every step on real checks: lint, tests, benchmarks, manual testing, independent code review, and PR monitoring. It stops when every acceptance criterion and quality gate passes, or when it hits a blocker it can name.
Run it with /backpressured:goal <description>, or by explicitly asking for a "backpressured" run. State your acceptance criteria in the prompt; if you don't, it infers and states them before starting.
One command from your terminal:
npx -y github:lucasfcosta/backpressured
Re-run it any time to update; it refreshes the marketplace and the plugin. Flags: --ssh (clone over SSH), --force (uninstall + reinstall, bypassing the version cache), --dry-run (preview).
Afterwards, run /reload-plugins or restart Claude Code.
Or, inside Claude Code:
/plugin marketplace add lucasfcosta/backpressured
/plugin install backpressured@lucasfcosta
/reload-plugins
Configuration is optional. Without any, the loop discovers your project's commands itself (package.json, Makefile, CI config, README).
To customize, drop a BACKPRESSURE.md at your project root with plain-English, project-specific instructions. It's authoritative: it layers on top of the defaults and wins on conflict. Common things to set:
react-best-practices, a design guide).You don't have to write it by hand. Run /backpressured:config and it detects your commands, surveys your skills and MCP servers, confirms each value with you, and writes the file. Re-run it any time to refresh.
Right now, humans spend a lot of time giving unnecessary feedback to an AI.
Most of that feedback could be replaced by series of automated checks, adversarial agentic reviews, and other forms of self-validation. That way, the agent can fix its own problems as it goes, without needing to stop and ask for help.
This skill set teaches the agent how to do that, and how to run a tight feedback loop that gates every step on real checks. The result is a more autonomous agent that can iterate faster and more reliably, reducing the need for human feedback.