self-improvement-pipeline
Nightly self-improvement pipeline for field agents. Agent reviews transcripts, identifies mistakes, auto-applies safe fixes, and sends proposals for risky changes to founders for approval. One job, no morning phase needed.
office-suite
Microsoft Office document manipulation (Word, Excel, PowerPoint, PDF). Use when: creating professional .docx documents, filling PDF forms, extracting tables from PDFs, building presentations, working with spreadsheets, or any task involving Office file formats. Triggers: Word, Excel, PowerPoint, PDF, .docx, .xlsx, .pptx, spreadsheet, presentation, slides, form, invoice, report, memo, letter, budget, financial model.
field-security
Security hardening for field agents serving external clients. Prompt injection detection, secret protection, client isolation enforcement, and security monitoring with HQ alerts.
skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
pptx
Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.
memory-retriever
Sub-agent memory retrieval system. Spawns a Sonnet sub-agent to search memory files, A2A logs, and learning entries — returns concise summaries without bloating the main session's context window.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
self-improvement
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
vulkn-software-fullstack
Field agent skill for operating full-stack infrastructure. The agent understands it has credentials and operates Supabase, Vercel, GitHub, and Railway independently — no human intervention needed.
vulkn-software-manager
HQ skill for provisioning full-stack infrastructure for new VULKN field agents. Creates GitHub repos, Supabase project, Vercel deploy, and Railway service — all via API. Sends individual keys to field agents via A2A.
memory-guardian
Layer 6: Memory integrity defense against persistent memory poisoning (Zombie Agent attacks). Scans memory files for injected instructions, tracks provenance, audits integrity. Passive monitor — zero friction on agent memory workflow.
nightly-report
Cron job for field agents to compile and send daily activity summaries to HQ. Tracks usage, sentiment, pain points, and improvement opportunities.
github-provision
Create frontend and backend repos from templates for a new client. Step 1 of vulkn-software-manager.
escalation-handler
Receive, triage, and resolve escalations from field agents. Route to founders when needed. Track resolution times.
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text or tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting or decrypting PDFs, extracting images, and performing OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.
supabase-provision
Create a Supabase project for a new client, retrieve keys, run migrations. Step 2 of vulkn-software-manager.
field-onboarding
First-run checklist for deploying a field agent to a new SMB client. Verifies profile, channels, connectivity, and cron setup. Domain knowledge handled by Johan.
agent-health
Monitor field agent activity and flag at-risk clients. Detect underuse, declining sentiment, and silent agents.
coherence-check
Closed-loop coherence verification for field agents. Detects behavioral drift, verifies learnings are applied, and compares outputs against brand profile and SOUL.md. Based on coherence debt theory (Mikkilineni 2026).
railway-provision
Create a Railway project and service linked to the backend repo with Supabase env vars. Step 4 of vulkn-software-manager.
docx
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of "Word doc", "word document", ".docx", or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a "report", "memo", "letter", "template", or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.
hive-mind
Collective intelligence system for VULKN multi-agent team. Enables cross-agent knowledge sharing, validated learnings, and institutional memory.
openclaw-sec
AI Agent Security Suite - Real-time protection against prompt injection, command injection, SSRF, path traversal, secrets exposure, and content policy violations
notion
Notion API for creating and managing pages, databases, and blocks.
risk-oracle
Pre-action risk assessment. Checks if proposed actions are similar to past corrections, giving the agent a 'sense of risk' before dangerous operations. Based on MemoryArena research showing agents fail to APPLY corrections even when stored.
cs-agent
Customer Service agent skills for HQ. Receives escalations and nightly reports from field agents, aggregates data, routes issues, and delivers insights to founders.
field-report
Nightly field agent reporting. Analyzes daily activity, generates structured reports with dashboard/insights/alerts, sends to HQ via A2A for founder briefings.
founder-briefing
Deliver concise daily and weekly summaries to founders. Distills field data into decisions, not dashboards.
agentic-learning
Unified learning system for persistent AI agents. Captures corrections, insights, and outcomes with stakes-gating, surprise filtering, auto-linking, outcome loops, and auto-promotion. Replaces self-improving-agent + standalone decision logger. Integrates with three-tier memory architecture as the Tier 3 engine.
manychat
ManyChat API for chatbot automation on Facebook, Instagram, WhatsApp, and Telegram. Send messages, manage subscribers, trigger flows, and set custom fields.
vercel-provision
Create a Vercel project linked to the frontend repo with Supabase env vars. Step 3 of vulkn-software-manager.
report-aggregator
Compile nightly reports from all field agents into aggregated dashboards. Identify trends, at-risk clients, and improvement opportunities.
Skill Creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
field-admin
Admin skills module for field agents deployed to SMB clients. Umbrella skill coordinating escalation, nightly reporting, and onboarding. Every field agent loads this.
VULKN Office
Professional documents for Mexican SMEs. USE THIS TOOL when the client requests invoices, quotes, contracts, reports, presentations, or any business document. Also for extracting data from received PDFs, filling forms, creating budgets in Excel, or generating pitch decks. Triggers: invoice, CFDI, quote, contract, report, presentation, estimate, RFC, SAT, document, Word, Excel, PowerPoint, PDF.
bjs-game
Strategic decision-making framework for BJS Labs. Use when: making significant business decisions, after completing major actions (close client, ship feature), weekly strategic review, or feeling uncertain about priorities. Tracks game state, win conditions ($300K/$500K/$1M ARR), and decision history. Triggers: strategy, decision, priority, should we, next move, game state.
consultation-oracle
Triggers and routing for field agents to consult the Frontier Lab when uncertain. Unlike escalation (crisis mode), this is for "I'm not sure" moments before they become problems.
xlsx
Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.
escalation
When and how field agents contact the customer service agent. Severity levels, auto-triggers, and message formatting for clean handoffs.
owner-playbook
Customer service playbook for handling business owner conversations. What to resolve alone, when to contact founders, response templates, and interaction logging.
collective-memory
Shared memory layer for agent collectives. Share learnings, search collective knowledge, synthesize insights. Includes PII protection.