Path-scoped. Loaded only when relevant.
Open a matching file → the right rules attach. Zero token cost the rest of the time.
What this pillar does for you.
Open a Button.tsx and component rules attach. Open a service and security rules attach. Zero cost otherwise.
The router reads short descriptions to decide. Full skill bodies only load when needed.
Long sessions get compressed packs. Halves token usage without losing the thread.
4 rules.
Each rule is a markdown file the AI loads when it edits a matching path. Click any to see related skills and where it's referenced.
Subagents leave a sub-2KB digest at phase boundaries. Downstream agents read the digest instead of re-reading the repo - cuts redundant token spend across the senior-eng orchestrator flow.
Read the active codebase-graph artifact (GitNexus or graphify) before falling back to broad Glob/Grep - replaces dozens of searches with one file read.
Versioned .traffic-one/ folder (product, stack, coding, security, schema, agent-log) survives across sessions, tools, and subagents - read first, then plan.
Architect-first, then frontend/backend parallel, then reviewer/tester parallel, shipper only on explicit deploy. Subagents own narrow write scopes and emit handoff digests.
4 skills.
Skills are intent-triggered workflows. The AI router loads them by description; bodies on demand.
Audit token consumption across rules, skills, hooks, and MCP servers; identify redundancies and produce a prioritized token-savings plan with concrete trims.
Create, refresh, or audit .traffic-one/ project memory: product, decisions, coding/security rules, schema.sql, deployments, known-issues, stack, mcp.json.
Cross-stack source-code asset audit: classifies every file, detects embedded third-party libs, and delivers a four-level verdict per module with HTML reports.
Orchestrate the architect -> frontend/backend -> reviewer/tester -> shipper team for multi-layer builds. Requires Performance level answer before implementation.
One plugin, eight pillars.
Every rule and skill is part of a system. Jump across pillars to see how they reinforce each other.
Use the AI you already pay for.
Add superpowers.
One plugin. Senior-engineer rules, security, a backend, and one-command deploys — inside the editor you already use.