Sudo AI

PDF task routing and recipes — extract text/tables/images, OCR scans, merge/split/rotate, watermark, passwords, fill forms, generate with reportlab. Condensed from grok's pdf skill; full doc in references/.

Explain any concept in plain words with one concrete analogy - clear enough for a smart child or a busy outsider, never condescending, never wrong.

Compress long text into a one-line takeaway plus the few points that matter - numbers and meaning-changing caveats preserved, nothing invented.

Spreadsheet deliverable doctrine — formulas not hardcoded values, zero formula errors via recalc, financial-model color/number conventions, template preservation, openpyxl vs pandas routing. Condensed from grok's xlsx skill; full doc in references/.

Safety rules and decision logic for media processing with ffmpeg/ffprobe via shell — inspect, convert, trim, merge, resize, compress, extract frames/audio, GIFs, subtitles, overlays. Condensed from grok's ffmpeg skill; full recipes in references/.

Image-generation doctrine for the media.* tools — when to build a visual with code instead of an image model, prompt-craft, reference-first handling of real people, factual grounding, verify-by-reading-back, and cross-image consistency. Ported and adapted from grok's imagine skill.

Diagnose and fix a bug through a disciplined reproduce, isolate, diagnose, fix, verify loop driven by evidence and one hypothesis at a time — not guesswork.

Design REST or RPC API endpoints with auth, versioning, error codes, and OpenAPI examples

Repair malformed JSON — trailing commas, unquoted keys, comments, encoding issues, and structural errors

Set up SSH tunnels for local port forwarding, remote forwarding, and key-based authentication

Review code for bugs, security vulnerabilities, style problems, and improvement opportunities

Visual design doctrine for slide decks — bold content-informed palettes, type pairing, varied layouts, and the assume-there-are-problems QA loop. Condensed from grok's pptx design guidance; pairs with the pptx.* tools. Full doc in references/.

Safety rules and decision logic for image processing with ImageMagick via shell — resize, crop, convert, watermark, composite, annotate, montage, batch. Condensed from grok's imagemagick skill; full recipes in references/.

Build, explain, and test regular expressions for any language or use case

Write unit and integration tests using Jest, Vitest, or pytest with proper mocking and coverage

Obtain and renew TLS certificates with Certbot and Let's Encrypt for nginx or standalone servers

Rewrite a blunt or rough email draft into a warm, friendly-but-professional version - tone and clarity improved, intent, facts, names, dates, and requests preserved exactly.

Branch, commit, merge, and release using Git best practices including conventional commits and rebase workflow

Write nginx server blocks for reverse proxies, static sites, HTTPS termination, and security hardening

Turn a raw transcript or scribble into decisions, action items with owners and dates, and open questions - so nothing agreed in the room evaporates afterwards.

Write pm2 ecosystem config files for Node.js process management with clustering, env vars, and log rotation

Scaffold docker-compose.yml files for common application stacks with volumes, networks, and health checks

Draft professional emails — requests, follow-ups, declines, announcements, and escalations

Improve clarity, concision, and flow of any prose — technical docs, blog posts, or business writing

Design a tool-using LLM agent with clean tool schemas, a control loop, explicit stop conditions, and guardrails.

Build an eval suite for an LLM feature with datasets, graders, and regression gates that block bad releases.

Translate human-readable schedules into cron expressions and explain existing cron syntax

Write systemd unit files to manage long-running services with proper sandboxing, restarts, and logging

Parse stack traces and error messages to identify root cause and suggest fixes

Manage Linux users, groups, sudo access, file permissions, and ownership

Scan a repository, its git history, and logs for leaked secrets and produce a credential rotation runbook.

Build a STRIDE-based threat model and attack-surface map for a feature or system so risks can be mitigated before they ship.

Run a periodic least-privilege access review that finds stale accounts, over-provisioning, and role drift, with owner attestation.

Write safe, performant SQL queries with parameterization, joins, and indexing guidance

Clean up and standardize markdown files — fix headings, lists, code blocks, and spacing

Run a pre-deploy verification covering code readiness, migrations, flags, config, observability, and a tested rollback before promoting to a shared environment.

Audit direct and transitive dependencies for known CVEs and produce a prioritized, reachability-aware remediation plan.

Drive a security incident through triage, contain, eradicate, recover, and blameless postmortem while preserving evidence.

Produce a CIS-style server and application hardening checklist with evidence and prioritized, exposure-ranked fixes.