ai-computer-ai

Plugin ai-computer-ai from marketplace ai-computer in hazemhagrass/ai-compute. Commands read from the repo's manifest — 30 days ago.

Skills (100)

• model-selection

  Use when choosing which AI model to use for a task. Match model class to the axis the task actually stresses, then measure on your own prompts.

• prompt-engineering

  Use when writing prompts for an LLM. Produce specific, delimited, example-driven prompts and parse model output defensively with a fallback path.

• smythos-sdk

  Use when code imports @smythos/sdk. Build SmythOS agents.

• cover-letter-generator

  Use when writing a cover letter for a job application. Say the thing the resume cannot, or send nothing.

• creative-resume-design

  Use when designing a visually distinctive resume. Typographic craft that reads well in print while a second parse-safe version feeds the ATS.

• infographic-resume

  Use when charts or skill graphics go on a resume. Build the visual form honestly: real axes, real quantities, evidence instead of skill-percentage bars.

• interactive-web-resume

  Use when shipping a resume as a web page. Build one semantic, fast, printable page whose printed PDF and web version never drift apart.

• interview-prep

  Use when preparing for a job interview. Rehearse a story bank of real situations so the preparation survives contact with the room.

• latex-resume

  Use when building a resume in LaTeX. Survive the ATS parser's text extraction while still reading well to a human.

• linkedin-profile-optimizer

  Use when optimising a LinkedIn profile for search. Make it findable in recruiter field search and credible to the human who opens it.

• portfolio-website-builder

  Use when building a portfolio site to get hired. Structure it around case studies with decisions and outcomes rather than a screenshot gallery.

• resume-storytelling

  Use when a resume reads as a job list, not a story. Turn duties into a narrative a reader infers correctly.

• salary-negotiation

  Use when negotiating a job offer or compensation. Get more of the package without putting the offer at risk.

• video-resume

  Use when an application wants a video introduction. Decide whether to skip it, or record one that does not hurt the candidacy.

• word-resume-optimizer

  Use when optimising a .docx resume for ATS parsers. Build linear Word documents whose text survives machine extraction and still reads well to a human.

• podcast-production

  Use when producing a podcast episode. Structure, record, edit, and publish so listeners can follow it and find it.

• social-media-scheduler

  Use when planning or scheduling social posts. Native-first drafts, a rolling calendar, honest repurposing, and no fake engagement.

• subtitle-finder

  Use when finding subtitles for media you already have. Match by moviehash first; inspect the container; verify sync before you keep.

• video-script-writer

  Use when scripting a video. Write for the ear, hook with substance, and keep one idea per segment.

• youtube-seo-optimizer

  Use when publishing on YouTube. Write titles and descriptions that serve viewers, and read CTR with retention.

• data-visualization-principles

  Use when building a chart. Pick the form from the question being asked, then refuse the distortions: truncated bars, dual axes, mean-only summaries, and colour used as decoration.

• powerbi-reports

  Use when building a Power BI model. Star schema, relationship cardinality, a real date table, measures over calculated columns, and the DAX contexts that decide whether the number is right.

• python-pandas-analysis

  Use when analyzing data with pandas. Avoid the silent correctness traps (chained assignment, dtype drift, merge row loss) and write fast, verifiable transforms.

• sql-for-analysts

  Use when writing analytical SQL. Get the number right: NULL semantics, join fan-out, grain, windows, half-open dates, reconciliation.

• tableau-dashboard-builder

  Use when building a Tableau dashboard. Get the aggregation, the LOD expressions, and the filter order of operations right, then reconcile the headline number against source before publishing.

• accessibility-audit

  Use when auditing a web UI. Test against WCAG AA by hand. Semantic HTML first, keyboard operability, real contrast ratios, correct focus behaviour, and ARIA only as a last resort.

• presentation-design

  Use when building slides or a deck. Lead with the conclusion, one idea per slide, and titles that assert a message instead of naming a topic.

• ci-cd-debugging

  Use when a pipeline fails or tests flake. Debug CI systematically: isolate flake vs red, repro, bisect the stage.

• docker-troubleshooting

  Use when a container won't start or a build breaks. Debug Docker systematically: image, layers, network, volume.

• kubernetes-debugging

  Use when a pod won't start, stays pending, or crashes. Systematic kubectl debugging.

• code-graph

  Use when navigating a codebase too big to grep. Builds and queries a real dependency graph.

• dotfiles-sync

  Use when syncing dotfiles across machines. Chooses bare-repo vs symlink manager, keeps host-specific config and secrets out, and bootstraps a new machine idempotently.

• env-doctor

  Use when a fresh clone breaks. Diagnose missing deps and env vars.

• git-workflow

  Use when committing, branching, or fixing git state. Enforces one-change commits, why-focused messages, pre-commit checks, and safe history recovery.

• regex-builder

  Use when writing a regex. Build it incrementally against real input, anchor it, and prove it cannot backtrack catastrophically.

• repo-snippet-extractor

  Use when a doc embeds code from a repo. Extract snippets by anchor, verify them in CI so they never drift.

• shell-history-alias-miner

  Use when typing one long command repeatedly. Mine shell history into aliases with collision checks.

• tmux-workspace

  Use when a terminal job must survive an SSH drop. Enforces named tmux sessions, scripted layouts, sane copy mode, and a portable .tmux.conf.

• api-design

  Use when designing or changing an HTTP API. Enforce boundary validation, informative status codes, branchable errors, tri-state patch fields, pagination, idempotency, secret masking, and additive versioning.

• api-integration

  Use when integrating a third-party API. Handle auth, retries, rate limits, and errors robustly.

• architecture-review

  Use when reviewing an architecture or design document. Force the author to state the optimised constraint, produce rejected alternatives, name failure modes, and prove an incremental rollout and rollback path.

• code-review-automation

  Use when CI bots or AI reviewers comment on PRs. Decide what to automate, gate on, and silence.

• code-review-checklist

  Use when reviewing a PR. Run fixed single-concern passes. Order them intent, correctness, security, tests, absences, then style, and verify each test fails without the change.

• code-review

  Use when reviewing a diff or pull request. Review in severity-ordered passes, judge against stated intent, and label every comment blocking/should-fix/nit.

• comment-removal

  Use when removing comments from a codebase en masse. Strip with parser-verified edits and keep functional directives.

• database-design

  Use when designing a schema or writing a migration. Enforce constraints in the database, reversible migrations, correct types for money and time, and deliberate indexing.

• debugging

  Use when chasing a bug or unexplained failure. Reproduce first, binary search the change set, input and code path, and fix the root cause instead of the symptom.

• frontend-architecture

  Use when structuring a React or Next.js frontend. Enforces server-first composition, seeded data, local state, and per-region loading and error boundaries.

• performance-profiling

  Use when something is slow. Measure first, fix the biggest contributor, stop at a stated target.

• refactoring-safety-checks

  Use when verifying a refactor did not change behaviour. Prove equivalence with a recorded baseline, a reference sweep, and a rollback plan before merging.

• refactoring

  Use when restructuring existing code. Hold behaviour still with tests, then apply one named transformation per commit and stop at a stated goal.

• site-test

  Use when running a repeatable QA pass on a live site. Browser + journey + code, safety-first, kept current with drift.

• sql-optimization

  Use when a query is slow. Find the bottleneck with EXPLAIN, fix indexes, rewrite the query.

• test-strategy

  Use when writing tests or planning coverage. Choose the level where the bug would actually be caught, test pure logic and risky branches first, and name tests after the rule they enforce.

• budget-tracker

  Use when building or fixing a personal budget. Track actuals before setting targets, separate fixed from variable from irregular, fund the irregular with sinking funds, and reconcile against the bank so the numbers stay real.

• investment-portfolio-analyzer

  Use when analyzing a portfolio's return and risk. Compute time-weighted vs money-weighted return, fee drag, concentration, and drawdown from your own data instead of trusting a broker summary screen.

• retirement-calculator

  Use when modelling a retirement projection. Fix real vs nominal, model sequence risk, output a range not a number.

• adguard-home

  Use when deploying network-wide DNS filtering. Set up AdGuard Home, point clients at it, stop bypass.

• lancache

  Use when setting up game caching for LANs or homelabs. Cache Steam/Epic/Origin downloads once, serve many.

• unraid

  Use when managing Unraid servers. Setup arrays, Docker, VMs, shares, troubleshoot parity/corruption.

• airtable-database-builder

  Use when designing an Airtable base. Model one table per entity, join with linked records, keep views as saved queries, and name the point where the base should become Postgres.

• obsidian-zettelkasten

  Use when an Obsidian vault stops paying off. Keep notes atomic and linked so you can write from them later.

• onenote-knowledge-base

  Use when OneNote is your knowledge base. Keep pages findable, shareable, and honest about what export loses.

• course-creation

  Use when designing a course or lesson. Design backwards from observable learner behaviour: objectives with verbs, one objective per lesson, practice before performance, and feedback that works without you in the room.

• spaced-repetition

  Use when building flashcards for spaced repetition. Write cards that force recall of one fact, and prune the deck before review debt buries you.

• context-file-audit

  Use when auditing always-loaded agent context files. Classify decay, verify every claim, change nothing before approval.

• skill-authoring

  Use when writing or editing an agent skill file. Enforces trigger-first descriptions, lessons over logs, and progressive disclosure.

• excel-dashboards

  Use when building an Excel dashboard or KPI sheet. Separate data from presentation, encode honestly, reconcile every headline number.

• excel-data-cleaning

  Use when cleaning a messy spreadsheet. Fix types, duplicates, dates and hidden characters on a copy, with row counts checked at every step.

• excel-financial-modeling

  Use when building a three-statement model or DCF. Separate inputs from calculations, prove the balance sheet balances.

• excel-formulas

  Use when writing or auditing Excel formulas. Avoid the silent wrong-answer traps (approximate VLOOKUP, fill-broken references, IFERROR masks, text-vs-number keys) and build sheets that stay correct when data moves.

• excel-macros-vba

  Use when writing or fixing an Excel macro. Kill Select/Activate, batch I/O arrays, restore app state, keep .xlsm trusted.

• google-sheets-advanced

  Use when a Sheet is slow or QUERY misfires. Covers Sheets-only behaviour Excel habits get wrong.

• pivot-tables

  Use when summarising data with a pivot table. Get the grand totals right, avoid silent undercounts, stale caches, and averaged averages.

• power-query-etl

  Use when a Power Query refresh is slow or breaks. Folding, staging, error rows, parameters that survive renames.

• slide-deck-designer

  Use when planning the content of a talk or a deck. Assertion-evidence slides, one idea per slide, slide count against runtime, rehearsal, and live-demo fallbacks.

• email-efficiency

  Use when an inbox is out of control. Triage in two passes, close loops in replies, and put follow-ups in a system instead of memory.

• academic-paper-writing

  Use when writing or revising a paper for peer review. Structure IMRaD, state contributions, survive the desk reject.

• citation-manager

  Use when managing a .bib file or bibliography. Verify every entry, dedupe on DOI, keep citations reproducible.

• grant-proposal-writing

  Use when writing or revising a grant proposal. Aims, budget, panel reading.

• literature-review
• research-data-analysis
• survey-design
• systematic-web-research
• truth-first
• api-security-audit
• authentication-audit
• cryptography-audit
• secrets-management-audit
• security-audit
• autonomous-task
• grill-me
• meeting-facilitation
• planning
• spec-first-development
• blog-post-writer
• copywriting-formulas
• newsletter-writer
• review-comment-phrasing
• technical-writing

Similar plugins