aemerson-commits
- 50 skills
- 0 followers
- 5 hours ago last updated
- ▌ PDF · aemerson-commitsRead, create, merge, split, OCR, watermark, encrypt, fill forms, or extract from PDF files. Trigger when user mentions .pdf or asks to produce one.
- ▌ DOCX · aemerson-commitsCreate, read, edit, or manipulate Word documents (.docx). Trigger on 'Word doc', '.docx', or formatted document requests (TOC, letterhead, templates). Not for PDFs, spreadsheets, or internal comms (use /internal-comms).
- ▌ XLSX · aemerson-commitsRead, create, edit, or convert spreadsheet files (.xlsx, .xlsm, .csv, .tsv). Trigger on spreadsheet references or tabular file tasks. Not for Word docs, HTML reports, or Google Sheets API.
- ▌ MCP Builder · aemerson-commitsBuild MCP servers for LLM-service integration. Python (FastMCP) or Node/TypeScript (MCP SDK).
- ▌ Canvas Design · aemerson-commitsCreate visual art (.png, .pdf) — posters, designs, static artwork. Trigger on art/design creation requests. Original designs only.
- ▌ Theme Factory · aemerson-commitsStyle artifacts (slides, docs, HTML pages) with 10 pre-set themes or generate custom themes on-the-fly.
- ▌ Internal Comms · aemerson-commitsWrite internal communications using company formats: status reports, leadership updates, newsletters, FAQs, incident reports, project updates.
- ▌ Algorithmic Art · aemerson-commitsAlgorithmic art via p5.js with seeded randomness. For generative art, flow fields, particle systems. Original work only.
- ▌ Slack Gif Creator · aemerson-commitsCreate animated GIFs optimized for Slack. Constraints, validation, and animation concepts.
- ▌ Evolve · aemerson-commits bundleEvolve — Continuous Learning Pipeline. Manages automatic behavioral pattern extraction from Claude Code sessions. Use to install hooks, run analysis, and graduate instincts.
- ▌ Wisdom · aemerson-commits bundleWeekly knowledge review — runs skill-audit, then analyzes instincts vs CLAUDE.md vs skills. Proposes new skills, skill enhancements, CLAUDE.md cleanup, and marks covered instincts. Triggered from /start-day on Fridays.
- ▌ Consult · aemerson-commitsGet an independent second opinion on a DECISION (not a diff) by writing a consult brief for a second model or agent with its own repo access — one that hasn't seen your reasoning chain. Use when a choice has real tradeoffs, when you've formed a view and want it attacked, or when a plan is about to be committed to. Triggers: 'get a second opinion', 'consult another model on this', 'is this plan right', 'what would a fresh read say'. NOT for reviewing a code diff (use your review/adversarial-review skill) and NOT for continuing your own work (that's a handoff).
- ▌ Handoff · aemerson-commitsWrite a disposable session-handoff doc to OS temp so a fresh session/agent can continue ONE in-flight task slice without re-deriving context. Triggers: 'hand this off', 'write a handoff', 'spin up a fresh session to continue X', or when context is getting large and you want a clean continuation. NOT for end-of-day notes (use /session-notes) or cross-session promises (use your memory/notes file).
- ▌
- ▌ Grill Me · aemerson-commitsInterrogate the user about their idea before any work begins. Works the design tree in rounds - asks the whole unblocked frontier at once with a recommended answer each, self-serves facts from the codebase. Produces an Intent Summary for /research-gate.
- ▌ Env Audit · aemerson-commitsAudit environment variables across all projects — detect missing secrets, compare code references against deployed config, and identify which env vars each context requires.
- ▌ Start Day · aemerson-commitsMorning startup: git pull, QMD update, session notes review, alerts check, ideas triage, and daily briefing
- ▌ Deploy All · aemerson-commitsParallel deploy all changed projects — detects changes, builds, deploys, watches CI, verifies live URLs. Use instead of manual per-project deploys. Triggers: 'deploy all', 'deploy everything', 'push and deploy'.
- ▌ Write Plan · aemerson-commitsWrite a structured implementation plan for multi-step features. Decomposes work into tasks with exact file paths, code examples, and verification steps. Use after /research-gate for complex features spanning 5+ files or multiple projects.
- ▌ HTML Slides · aemerson-commitsHTML Slides — Create polished, interactive HTML presentation slides from a single prompt. Output is a self-contained HTML file. Triggers: 'create a presentation', 'make slides', 'build a deck'.
- ▌ Review Impl · aemerson-commits bundleAdversarial implementation review. Dispatches independent reviewer agents to verify spec compliance and code quality after feature implementation. Use after completing a multi-step feature, before committing or deploying.
- ▌ Skill Audit · aemerson-commitsRun a comprehensive health audit across all skills — checks staleness, safety guards, cross-skill consistency, context efficiency, step compliance, learning pipeline, externally-synced skills, and unversioned extension surfaces. Use when you want to verify skill quality, after bulk skill edits, or as a periodic maintenance check.
- ▌ Schema Check · aemerson-commitsVerify database schema, KV keys, and env bindings before writing queries or endpoints. Use BEFORE any SQL, KV operations, or config changes. Triggers: 'check schema', 'what columns does X have', or proactively when writing DB/KV code.
- ▌ Triage Ideas · aemerson-commitsProcess Obsidian Ideas.md inbox — route ideas to project roadmaps. NOT for direct roadmap edits or reviews.
- ▌ Worker Build · aemerson-commitsFactory for building Cloudflare Workers — email reporters, cron jobs, API monitors. Parallel agents: scaffold, admin UI, secrets config, deploy + verify.
- ▌ Gh Repo Recon · aemerson-commitsRead a GitHub repository you haven't cloned — its code, its issues and PRs, or its security posture — through the `gh` API instead of cloning it. Use when evaluating a dependency or platform before adopting it, tracing how another repo implements something, auditing rulesets / branch protection / deploy keys / Dependabot alerts, or answering 'who merged what' on a repo whose files you don't need. Triggers: 'look at <owner>/<repo>', 'how does X implement Y', 'audit our repo settings', 'check branch protection', 'who merged these PRs'. NOT for the repo you already have checked out (use Grep/Read) and NOT for CI status on your own PR (use `gh pr checks`).
- ▌ Research Gate · aemerson-commitsResearch-before-code gate. Trigger PROACTIVELY for features spanning 3+ files, external APIs, or unknown constraints. Trigger words: 'new view', 'integrate with', 'build a system'. Skip single-file fixes or known-constraint tasks.
- ▌ Session Notes · aemerson-commits bundleEnd-of-session notes. Two modes — cheap `checkpoint` for mid-day, full close-out at true end of day. Triggers: 'close out', 'wrap up', 'done for the day', 'checkpoint'. NOT for reading past notes.
- ▌ Skill Creator · aemerson-commitsCreate, modify, test, and optimize skills. Use for new skill creation, eval testing, performance benchmarking, or description optimization.
- ▌ Isolation Test · aemerson-commitsTeam C diagnostic — isolation testing when multiple factors contribute to a persistent issue. Tests components independently to find interacting failures.
- ▌ Webapp Testing · aemerson-commits bundleToolkit for interacting with and testing web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
- ▌ Worktree Guard · aemerson-commitsSafe worktree merge — diffs each changed file against dev HEAD before copying, flags conflicts. Use BEFORE copying worktree files to the main tree. Triggers: 'merge worktree', 'copy from worktree', after any worktree agent completes with changes.
- ▌ Deep Root Cause · aemerson-commitsTeam A diagnostic — deep root cause analysis when the obvious fix was wrong. Traces data flow, audits assumptions, builds reproduction.
- ▌ Design Showcase · aemerson-commits bundleBuild a side-by-side visual comparison of 3-7 design variants for a component or view. Dispatches a single Opus agent that produces a self-contained HTML showcase, serves it locally, and lets the user pick a direction. Use when the user wants to explore design options before committing to an implementation.
- ▌ Frontend Design · aemerson-commitsDesign system guide — CSS tokens, Tailwind v4, shadcn/ui components, theming, and UI conventions for consistent dark-themed dashboard interfaces.
- ▌ Model Selection · aemerson-commitsModel + thinking-effort orchestration. Reference when dispatching subagents (Agent tool or workflow agent() calls) and when setting session effort. Sonnet is the subagent floor; the main-loop model is the expensive decision-maker; effort is routed per task instead of maxed globally.
- ▌ Verify Complete · aemerson-commitsEvidence-based completion check before claims persist to persistent state, session notes, or roadmap. Use BEFORE writes that say items are done, BEFORE answering 'how much is done?', or before any multi-item completion claim.
- ▌ Audit Components · aemerson-commitsAudit component health after code changes — verify imports, props, and shared utility consistency across all projects.
- ▌ Full Stack Build · aemerson-commits bundleOrchestrate agent teams to build features spanning frontend + backend — API design, D1 migrations, React components, integration wiring, and verification. Use for new views with endpoints.
- ▌ Full Stack Trace · aemerson-commitsTeam B diagnostic — systematic layer-by-layer trace when the fix was in the wrong layer. Frontend -> API -> Backend -> Infrastructure.
- ▌ Persistent Issue · aemerson-commitsEscalating diagnostic router for persistent issues. Classifies, dispatches Team A-E, auto-escalates in cascade mode. Triggers: 'not fixed', 'still broken', 'same error'.
- ▌ Pre Merge Review · aemerson-commitsComprehensive pre-merge review — security pentesting, performance, scalability, UX consistency, and code hardening — then works the findings through to landed fixes on a branch. Run before merging dev to main to catch issues before production. The deliverable is a merge-ready branch, not a findings list: after synthesis the skill verifies each finding, triages apply/flag/outward-facing, fixes on a branch cut from the reviewed SHA with a regression test that fails on the unfixed code, and re-runs the full pre-flight. Triggers: 'pre-merge review', 'review before merge', 'full code review for merge', 'fix the review findings'. Soft-default: when planning a commit or merge that touches 3+ top-level projects, suggest running this skill before pushing — the cross-project blast radius warrants a review even if no individual change looks risky.
- ▌ Throwaway Script · aemerson-commitsPattern for writing & executing throwaway .mjs Node scripts that need .env secrets — covers location choice, ES-module form, Windows + Git Bash path quirks, and noise filtering. Use for ad-hoc API probes, polling, calibration runs, smoke verification.
- ▌ Debug Collaborate · aemerson-commitsMulti-agent collaborative debugging — parallel hypothesis generation and testing. Use when a solution didn't work, a bug persists, or the user reports 'still happening' / 'not fixed' / 'didn't work'.
- ▌ Propagate Feature · aemerson-commitsReplicate features across projects with independent codebases — diff, adapt, apply. For cross-project propagation when projects have component copies instead of shared imports.
- ▌ Regression Bisect · aemerson-commitsTeam E diagnostic — regression hunting for features that stopped working. Git bisect, side effect analysis, rollback verification.
- ▌ Temporal Forensics · aemerson-commitsTeam D diagnostic — temporal analysis for intermittent issues. Hunts race conditions, stale state, cache timing, and async ordering.
- ▌ Data Reconciliation · aemerson-commitsTrace a data item through your pipeline layers to find where values diverge. For debugging stale/wrong data, not code bugs.
- ▌ Cascade Orchestration · aemerson-commitsReusable agent team cascade patterns — parallel dispatch, sequential pipelines, fan-out/fan-in, and cross-project propagation
- ▌ Design Reverse Engineer · aemerson-commits bundleReverse-engineer a design system from a given URL — typography, color palette, spacing, component patterns, micro-interactions — and produce a Claude-ready design spec that can guide future UI work. Use when building a new view and you want to anchor it in a proven design system rather than generic AI output.