psc-comet

Convert PDF/EPUB library to Markdown and generate Obsidian MOC notes

Hook-based compaction suggestions at logical task boundaries

Context window management — track spend, decide when to compact, preserve state

Session-start orientation — loads context, surfaces learnings, confirms registry

Quality and semantic review — catches what automated tools miss

Planner → Architect → Critic deliberation loop — produces a formally validated ADR

Dual-dimension document quality gate — presentation rubric + hallucination grounding check

Model selection guidance — match task complexity to model capability

Scan task for high-cost branch points, confirm which get deep reasoning, execute with gates active

Resume an interrupted mid-task session — reconstructs state from context files and picks up at the right step

Full feature workflow — deep-interview → prd → consensus-plan → tdd → code-review → security-gate

Plan-mode enforcement — required before edits affecting 3+ files, cross-domain work, or security-sensitive changes

Session-end protocol — collect learnings, update context, commit session state

Framework for decomposing agent-driven tasks into independently verifiable units

Extract instinct candidates from the current session and add confirmed ones to the instinct store

Eval-driven development with pass@k metrics and three grader types

Capture codebase-specific knowledge as a reusable skill — passes 3-point quality gate before writing

Mid-session quick capture — save an insight before the context moves on

Auto-populate context/project.md by reading the current codebase

In-session memory tagging — mark content for recall later in the same session

Rules of Engagement authorization gate — required before any security operation, autonomous scanning, or action with real-world effects outside the local codebase

Pre-deploy security assessment — validates against OWASP Top 10 and project standards

Prevent algorithm confusion attacks when validating JWTs — RS256 tokens accepted as HS256

UI/UX design intelligence for web and mobile. Includes 50+ styles, 161 color palettes, 57 font pairings, 161 product types, 99 UX guidelines, and 25 chart types across 10 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, and HTML/CSS). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, and check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, and mobile app. Elements: button, modal, navbar, sidebar, card, table, form, and chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, and flat design. Topics: color systems, accessibility, animation, layout, typography, font pairing, spacing, interaction states, shadow, and gradient. Integrations: shadcn/ui MCP for component search and examples.

Framework for designing quality agents with proper action space and contracts

Test specifically for AI-introduced regressions that repeat without tests

Audit animations and transitions for motion accessibility, performance safety, and design intent. Enforces prefers-reduced-motion compliance and blocks layout-triggering transitions.

Parallel agent swarm — decomposes work into independent units, spawns isolated workers, tracks PRs via fan-in

Performance measurement with Core Web Vitals thresholds — baseline capture, regression detection, optional PageSpeed Insights

Load and validate brand assets before any design work — logo, colors, typography, voice. Gates UI and marketing work behind confirmed brand context.

Post-deploy production monitoring — watches live app for errors, regressions, and failures after deployment

Named decision point for long-running tasks — PIVOT, REFINE, or PROCEED

4-phase systematic codebase onboarding with reconnaissance and artifact generation

Define component API, states, variants, and ARIA requirements before writing a line of implementation code. Gates component work behind a confirmed spec.

Token audit across agents, skills, and MCP with cost optimization guidance

Cross-session instinct learning with confidence-based promotion and global scope

Chief Security Officer mode — multi-phase security audit covering OWASP, STRIDE, secrets, CI/CD, and LLM attack surfaces

Structured debugging protocol — hypothesis-driven investigation with evidence tracking

Socratic problem exploration — identifies critical unknowns and gates on ≤20% ambiguity before proceeding

6-step multi-source research workflow with inline citations and synthesis