andrem-sec/psc-comet

library-pipeline

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

Langue du texte source : anglais

strategic-compact

Hook-based compaction suggestions at logical task boundaries

Langue du texte source : anglais

token-budget

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

Langue du texte source : anglais

heartbeat

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

Langue du texte source : anglais

code-review

Quality and semantic review — catches what automated tools miss

Langue du texte source : anglais

consensus-plan

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

Langue du texte source : anglais

doc-review

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

Langue du texte source : anglais

model-router

Model selection guidance — match task complexity to model capability

Langue du texte source : anglais

reasoning-gates

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

Langue du texte source : anglais

resume

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

Langue du texte source : anglais

feature-pipeline

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

Langue du texte source : anglais

plan-first

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

Langue du texte source : anglais

wrap-up

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

Langue du texte source : anglais

agentic-engineering

Framework for decomposing agent-driven tasks into independently verifiable units

Langue du texte source : anglais

reflect

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

Langue du texte source : anglais

eval-harness

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

Langue du texte source : anglais

learner

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

Langue du texte source : anglais

lesson-gen

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

Langue du texte source : anglais

project-scan

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

Langue du texte source : anglais

remember

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

Langue du texte source : anglais

roe

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

Langue du texte source : anglais

security-gate

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

Langue du texte source : anglais

jwt-algorithm-confusion-prevention

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

Langue du texte source : anglais

ui-ux-pro-max

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,…

Langue du texte source : anglais

agent-harness-construction

Framework for designing quality agents with proper action space and contracts

Langue du texte source : anglais

ai-regression-testing

Test specifically for AI-introduced regressions that repeat without tests

Langue du texte source : anglais

animation-safe

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

Langue du texte source : anglais

batch

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

Langue du texte source : anglais

benchmark

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

Langue du texte source : anglais

brand-context

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

Langue du texte source : anglais

canary

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

Langue du texte source : anglais

checkpoint

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

Langue du texte source : anglais

codebase-onboarding

4-phase systematic codebase onboarding with reconnaissance and artifact generation

Langue du texte source : anglais

component-spec

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

Langue du texte source : anglais

context-budget

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

Langue du texte source : anglais

continuous-learning-v2

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

Langue du texte source : anglais

cso

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

Langue du texte source : anglais

debug-session

Structured debugging protocol — hypothesis-driven investigation with evidence tracking

Langue du texte source : anglais

deep-interview

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

Langue du texte source : anglais

deep-research

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

Langue du texte source : anglais