davesnx

@davesnx source repo

53 skills
0 followers
3 hours ago last updated

▌
Plannotator · davesnx bundle
Reference for using the Plannotator CLI: plan review, code review, annotating files, URLs, folders, and running local apps, annotating the last assistant message, browsing archived plan decisions, and exporting or sharing Guided Reviews. Invoke when asked to use Plannotator for anything not covered by a more specific plannotator-* skill.

0
installs
▌
Supabase Postgres Best Practices · davesnx bundle
Postgres best practices (schema design, migrations, RLS, indexes, functions, pgvector, pg_cron) for any Postgres database. Load before writing or changing anything in a Postgres schema or query, or when diagnosing slow queries, locking, or connection issues.

0
installs
▌
Opam · davesnx bundle
opam is the package manager for ocaml packages. This skill teaches you how to use it

0
installs
▌
How · davesnx bundle
Explain or research how something works in this codebase by exploring code and producing a clear architectural explanation. Use for "how does X work", "research X", "trace X", or "where is X used" when X refers to the current codebase. Optionally critique the architecture for issues.

0
installs
▌
Tdd · davesnx bundle
Use only when the user explicitly asks for TDD, test-first implementation, red-green-refactor, or integration tests as test-first work. Do not use for generic implementation, tests without a test-first request, bug fixes, or domain-modeling artifacts. If the user explicitly asks for both test-first work and domain modeling, load both skills.

0
installs
▌
Eli5 · davesnx
Give a short visual explanation in chat by default. Create one self-contained HTML file only when the user explicitly requests HTML, a presentation, or an interactive explanation. Use ONLY when the user explicitly asks to use the eli5 skill.

0
installs
▌
Debug · davesnx bundle
Diagnosis loop for bugs and performance regressions. Use when the user explicitly asks to diagnose or debug, or provides a concrete runtime failure, shares a log file or measured regression.

0
installs
▌
Enpass · davesnx bundle
Use when the user asks to list local Enpass credential titles, retrieve or copy an Enpass password, or provide an Enpass secret to a command. Uses the safe local `enpass` wrapper and prevents broad substring matches or password exposure in agent output. Works on macOS via the keychain and on Linux only when `ENP_PIN` is set; never calls `enpass-cli` directly.

0
installs
▌
Github · davesnx bundle
GitHub and Git work: validated commits, GitHub Actions CI failures, non-trivial gh commands, and merge/rebase/cherry-pick conflict resolution. Load before every `git commit`. Use for 'commit', 'write a commit message', 'commit changes', GitHub PR inspection, unresolved review threads, GitHub CI logs, 'fix merge conflicts', 'resolve conflicts', or a conflicted branch.

0
installs
▌
Recall · davesnx bundle
Reconstruct recent working context from chat history, repository state, and shared records like issues and incidents. Use for 'recall my work on X', 'catch me up', 'where did I leave off', 'what did we decide', 'resume work on X', or to summarise past conversations.

0
installs
▌
Triage · davesnx bundle
Move issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.

0
installs
▌
Unslop · davesnx bundle
Edit or review prose to remove AI-generated writing patterns. Use only for writing, documentation, copy, reports, or messages when the user asks to unslop or deslop text, remove AI tone, or clean up prose. Do not use for source code.

0
installs
▌
Council · davesnx bundle
Run a subagent council to investigate one problem from multiple perspectives and models, compare findings, and produce a recommendation. Use for a council, second opinions, multiple agents or models on one question, red-team/blue-team comparison, or deciding between competing approaches.

0
installs
▌
Reflect · davesnx bundle
Spawn three parallel review subagents over the active transcript, surface learnings, and route each to a concrete edit on an existing skill. Use when the user says reflect.

0
installs
▌
To Spec · davesnx bundle
Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

0
installs
▌
Variate · davesnx bundle
Puts real design variations of one project file on the user's own localhost, behind a small card that flips them with arrow keys, then keeps listening for their verdict. Use to compare design directions or alternatives. Works in an empty folder or a real project, on any stack. Not for a single change with one right answer, a refactor, or a non-visual edit: just edit the file for those.

0
installs
▌
Grilling · davesnx bundle
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, says "grill me" or "grill with docs", or uses another grill trigger phrase.

0
installs
▌
Architect · davesnx bundle
Design production modules, interfaces, seams, types, signatures, and dependency structure before implementation. Use ONLY when the requested result is a production code shape or integration contract, such as 'architect this' or 'design this module'. Do not use for throwaway HTML experiments that test state, transition, policy, scheduling, or algorithm behavior.

0
installs
▌
Blog Post · davesnx bundle
Use when the user wants to collect blog notes, write, co-author, improve, or review a blog post or article. Trigger on a title, topic, outline, rough draft, complete post, 'write a post about,' 'improve this post,' 'blog draft,' or 'editorial feedback.' Supports notes-only collection or interview-first, section-by-section drafting with an unslop pass and fresh-reader testing.

0
installs
▌
Create Pr · davesnx bundle
Use when asked to open, create, submit, or prepare a pull request. Checks the branch and required validation, delegates title/body writing to write-pr-body, and creates a draft PR only with explicit publishing authorization. For title/body text alone, use write-pr-body.

0
installs
▌
Librarian · davesnx bundle
Cache and refresh remote git repositories under ~/.librarian, grouped by host/org/repo, so future references can reuse a local copy. Use this skill when the user points you to a remote git repository as reference or you encountered a remote git repo through other means.

0
installs
▌
Setup Mlx · davesnx
Install and configure mlx, the JSX syntax for OCaml, in a dune project. Use when the user asks to set up, enable, or add mlx or JSX to an OCaml project, when a .mlx file exists but dune-project has no mlx dialect, or when mlx-pp, ocamlmerlin-mlx, or ocamlformat-mlx is missing.

0
installs
▌
X Writing · davesnx bundle
Write, edit, or analyze posts and threads for X, including audience fit, supplied analytics, and verified recommendation-system constraints. Use for X-content requests and diagnosis. Treat algorithm details as constraints and never promise reach. Do not use for developer documentation, which belongs to technical-docs, or for independent evidence gathering and cited technical findings.

0
installs
▌
Impeccable · davesnx bundle
Use for frontend visual design and UX work, including layout, styling, accessibility, responsive behavior, themes, typography, color, motion, interaction, UX copy, and design systems. Not for React implementation or performance work alone; use vercel-react-best-practices for that. Load both only when the request explicitly includes visual design and React performance work.

0
installs
▌
Loop On CI · davesnx bundle
Inspect, plan, fix, and monitor failing CI across GitHub Actions, Buildkite, and other attached providers. Use when a user asks to debug CI, summarize a red build or PR, propose a CI fix plan, fix failing checks, retry a flaky job, or watch checks until they pass. Accepts a PR, branch, commit, build URL, or provider-specific build identifier.

0
installs
▌
Mlx Syntax · davesnx
Syntax guide for mlx, the JSX syntax for OCaml. Use when writing, reading, or reviewing a .mlx file, when the user asks for JSX or HTML-like elements in OCaml, or when a function must be callable as an mlx tag.

0
installs
▌
Post Draft · davesnx bundle
Fill the gaps in a sancho.dev blog draft by interviewing the author in rounds, then writing the missing prose into the draft using VOICE.md as a contextual reference. Use when a draft under src/content/ has a SKILLS block with state stub, outline or partial, or when the user asks to finish a draft, fill what is missing in a post, or be grilled about a post.

0
installs
▌
To Tickets · davesnx bundle
Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.

0
installs
▌
Sync Skills · davesnx bundle
Sync shared agent skills between the local Mac and nspawn through the setup Git repository. Use when the user says "sync skills", "push skills", "send skills to nspawn", "update nspawn skills", or asks to synchronize OpenCode, Claude, Codex, or Agent Skills across these machines. On local, commit and push skill changes, then pull them on nspawn. On nspawn, commit and push skill changes only.

0
installs
▌
Verify This · davesnx
Verify a claim with fresh local evidence: restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE.

0
installs
▌
Auto Improve · davesnx bundle
Use proactively when a host requests an auto-improve checkpoint, after repeated corrections or tool failures, or when a reusable workflow emerges. Propose improvements to skills, hooks, scripts, or agent rules and ask for approval before editing. Limit automatic reviews to one per session. Also use when explicitly asked to auto-improve. Not a general code audit or routine cleanup.

0
installs
▌
Autoresearch · davesnx
Set up and run a bounded, measured experiment loop for an optimization target. Use when asked to start autoresearch, optimize a metric, or run experiments.

0
installs
▌
Skill Vendor · davesnx bundle
Keep the skills this repository publishes for other people, such as the mlx skills in ocaml-mlx/skills, in step with their upstream repository through the skill-vendor command. Use when the user says "vendor skills", "skill-vendor", "pull the mlx skills", "push the mlx skills upstream", or asks whether agents/skills/VENDOR is in sync. Not for moving skills between the Mac and nspawn; that is sync-skills.

0
installs
▌
Ssh Transfer · davesnx bundle
Transfer files or directories between the local Mac and nspawn or another SSH host with ssh-transfer. Use for SSH uploads, downloads, copying remote artifacts to the Mac, or sending local files to nspawn. Not for publishing public download URLs or syncing agent skills through Git.

0
installs
▌
Comment Purge · davesnx bundle
Use only for an explicit user request to clean up code comments or run Comment Purge, even on hosts that ignore disable-model-invocation. Delete unnecessary comments, review exceptions, and fix accepted in-scope causes. Not for general code review or refactoring.

0
installs
▌
Skill Creator · davesnx bundle
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.

0
installs
▌
Write Pr Body · davesnx bundle
Use when writing or editing a pull request title or body. Drafts concise, evidence-grounded text from the final branch diff, with proportional design context, code examples, visual comparisons, and benchmark tables. Does not publish PRs, run checks, or write commit messages.

0
installs
▌
Writing Shape · davesnx bundle
Shape a fixed pile of raw material (notes, transcripts, drafts) into an article, paragraph by paragraph, without adding new claims. Companion to blog-post and unslop.

0
installs
▌
Code Standards · davesnx
Shared code quality bar for complexity, boundaries, failures, state and type models, modules, tests, and operations. Load when a change alters one of those areas, or when simplify or code-review needs the shared standards.

0
installs
▌
Diagram Design · davesnx bundle
Create branded diagrams (architecture, flowcharts, sequence, ER, org charts, Gantt, and 30+ other types) as standalone HTML/SVG/PNG, redraw draw.io or Mermaid sources, or apply brand tokens from a website.

0
installs
▌
Playwright CLI · davesnx bundle
Automate browser interactions, test web pages and work with Playwright tests. Use it to open, control, or inspect the user's Brave browser through CDP.

0
installs
▌
Technical Docs · davesnx bundle
Write, edit, or review developer documentation: READMEs, RFCs, PRDs, tutorials, how-to guides, reference pages, explanations, runbooks, and architecture docs, including docs from supplied facts. PR titles and bodies belong to write-pr-body. Do not use for open-ended evidence gathering, source conflict analysis, or X posts, which belong to x-writing. Do not use for commit messages, product UI copy, or code review findings.

0
installs
▌
Domain Modeling · davesnx bundle
Use only when the user explicitly asks to model a domain, establish ubiquitous language, create or update a glossary or CONTEXT.md, map contexts, or create or revise an ADR. Do not use for TDD, test-first implementation, red-green-refactor, or integration tests alone. If the user explicitly asks for both domain modeling and test-first work, load both skills.

0
installs
▌
Logic Prototype · davesnx bundle
Build throwaway, self-contained, drivable HTML artifacts to answer state-model, transition, policy, scheduling, or algorithm questions. Use ONLY when the requested result is a disposable interactive logic experiment used to test behavior before production design or implementation; do not use for production modules, interfaces, types, seams, or architecture contracts.

0
installs
▌
Plannotator Review · davesnx bundle
Open Plannotator's browser-based code review UI for the current worktree or a pull request URL, then act on the feedback that comes back.

0
installs
▌
Writing For Agents · davesnx bundle
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.

0
installs
▌
Simplify · davesnx bundle
After working on code, ensure the changes contain only the minimal, idiomatic edits. Simplify or refactor existing code while preserving its required behavior, and remove AI-generated slop introduced by you. Use for explicit cleanup, refactoring, removal of unnecessary complexity, or a request to use Simplify

0
installs
▌
Plannotator Annotate · davesnx bundle
Open Plannotator's annotation UI for a markdown file, plain-text config file (.yaml, .json, .toml, .ini, .csv, .log, …), HTML file, URL, or folder and then respond to the returned annotations.

0
installs
▌
Execute Codebase Plan · davesnx bundle
Review, execute, reconcile, or publish an existing codebase audit plan. Use when the user points to an existing plan and asks to review, implement, reconcile, or publish it. Source edits happen only in an isolated worktree; never merge or push without authorization, and never touch the main worktree. Use improve-codebase-architecture instead to discover improvements or create a new audit.

0
installs
▌
Extract Design System · davesnx bundle
Extract design tokens (colors, fonts, spacing, radius, shadows) from a public website URL using the extract-design-system CLI and write starter token files for a new project. For repeated patterns inside your own local codebase, use the impeccable skill's extract mode instead.

0
installs
▌
Code Review · davesnx bundle
Review local changes, branch diffs, commits, or pull requests for correctness, security, regressions, maintainability, standards, spec compliance, and blast radius. Use for "review", "code review", "review PR", "deep review", "adversarial review", "thermonuclear", "blast radius", or "review and ship". Accepts an optional fixed point, PR number, or PR URL.

0
installs
▌
Vercel React Best Practices · davesnx bundle
Use for explicit React or Next.js performance work involving rendering, data fetching, bundles, RSC, hydration, or measured regressions. Do not use for visual design, layout, styling, accessibility, or general component edits; use impeccable for those. Load both only when the request explicitly includes visual design and React performance work.

0
installs
▌
Improve Codebase Architecture · davesnx bundle
Audit a codebase across correctness, security, performance, tests, and architecture; render vetted findings as a local HTML review and write implementation plans for another agent. Use for "improve this codebase", "audit this repo", or "what should we build next". Read-only on source. Do not use it to review, execute, reconcile, or publish an existing plan; use execute-codebase-plan for that.

0
installs

53 published skills