dowdiness

Error handling conventions for MoonBit projects. Use when designing error types, choosing between abort/fail/raise, writing catch blocks, defining FFI boundaries, or reviewing error handling patterns. Triggers on: error handling, abort, fail, raise, catch, Result, error types, fallible functions, boundary safety, error recovery.

Use when reviewing a MoonBit/Canopy change or pull request before merge and multiple independent review perspectives are useful

Use when writing or reviewing MoonBit code against the `dowdiness/incr` reactive library (v0.14.x+) — building Inputs/Deriveds/ReachableDeriveds, attaching long-lived derived cells with `Watch`/`Observer`, adding microbenchmarks, or wrapping a reactive pipeline in a struct. Catches recurring idiom misses (inside-vs-outside read semantics, GC-anchor `Watch`/`Observer`, `Type::Type` constructor naming, defensive copies).

Use when creating or editing a GitHub PR/issue body, comment, or commit message from a shell, or when writing a grep/rg search pattern — especially if the Markdown or pattern contains backticks, `$()`, `>`, or other shell metacharacters that could be expanded before the tool sees them.

Use when creating, removing, or working across git worktrees or submodules — before force-pushing, before deleting a worktree, after `git worktree add`, when merging or deleting branches with `gh pr merge`/`gh pr close`, when bumping a submodule pointer, when merging a stack of dependent PRs, or when another agent/process may be operating on the same checkout concurrently.

Use when writing MoonBit code involving derive(Debug) on container fields, a trait method whose name matches a callable field, JS-target-only public APIs or JS numeric formatting, `guard` early-exit syntax, package visibility (`pub` vs `pub(all)`) on mutable-container fields, or `moon fmt` touching a comment-only or multi-line closure body — silent-failure compiler/formatter behaviors that `moon check` and `moon fmt` don't flag.

Use when the user asks whether a PR is good/beneficial for the project, wants long-term (6–24 month) impact, architectural evaluation, technical-debt analysis, project-direction alignment, research/innovation value, a senior-maintainer/architect verdict, or follow-up issues derived from that analysis. NOT for ordinary bug-finding, security, or pre-merge correctness review (use code-review), and not for merging (use merge-pr).

End-of-session ritual: update memories for completed work, draft a self-contained next-session prompt, and report what's safe to drop from conversation context. Use when concluding a work session and preparing to /clear. Usage: /handoff [next-target-hint]

Tracks deprecated MoonBit syntax to avoid generating invalid code. Reference this before writing MoonBit code. Auto-maintained when deprecated patterns are discovered.

Guide to solving the Expression Problem in MoonBit using Finally Tagless encoding, two-layer architecture (tagless + concrete AST), polymorphic trait methods, and related patterns. Use when designing extensible data types, adding new variants or operations to existing code, working with Finally Tagless, Object Algebras, open recursion, or discussing extensibility trade-offs in MoonBit.

Use BEFORE any performance optimization in MoonBit. Requires reproducing the claimed bottleneck in an isolated microbenchmark before designing a solution. Triggers on "optimize", "performance", "bottleneck", "slow", "speed up", or any TODO item citing millisecond costs. Do NOT skip this for "obvious" optimizations.

Execution discipline for boundary-crossing MoonBit refactors: splitting packages with language-enforced isolation via `internal/`, extracting modules behind a `pub using` facade, splitting files safely, and pinning invariants with property tests before structural change. Use when the task is "split this package", "extract these files into an internal package", "verify the public API didn't change", or "set up tests before refactoring". Pairs with `moonbit-refactoring`, which covers *what* to refactor toward.

Reference guide for effective trait usage in MoonBit's Self-based trait system (no trait type parameters, no associated types, but polymorphic trait methods are supported in v0.10+). Use when writing MoonBit traits, designing APIs with traits, or when the user asks about trait patterns like endomorphisms, capability traits, polymorphic methods, callback-based iteration, trait multiplication, newtypes, visitor pattern, or defunctionalized associated types in MoonBit.

Set up MoonBit project instructions and agent configuration for coding assistants. Use when bootstrapping agent guidance, validation commands, hooks, project conventions, or shared MoonBit context for Codex, Claude Code, or other agent environments.

Error handling conventions for MoonBit projects. Use when designing error types, choosing between abort/fail/raise, writing catch blocks, defining FFI boundaries, or reviewing error handling patterns. Triggers on: error handling, abort, fail, raise, catch, Result, error types, fallible functions, boundary safety, error recovery.

Repo maintenance for MoonBit projects. Five subcommands: default (full audit-and-fix pipeline), check (fast read-only health check), fix (full check + auto-fix only), triage (project direction + branch pruning), release (pre-release prep). Use at session start, before commits, weekly for direction, and before releases.

Use this skill when the user asks to save, remember, recall, or organize memories. Triggers on: 'remember this', 'save this', 'note this', 'what did we discuss about...', 'check your notes', 'clean up memories'. Also use proactively when discovering valuable findings worth preserving.

Rabbita binding and idiom reference. Use BEFORE designing, implementing, or reviewing any code that uses @sub, @cmd, @html, @dom, @http, or that authors or modifies a rabbita binding (e.g. lib/rabbita_codemirror). Requires reading the local rabbita repository's authoritative docs before answering. Triggers on 'rabbita', '@sub', '@cmd', '@html', '@dom', '@http', 'custom_sub', 'custom_cmd', 'suberror', 'Emit', 'Sub binding', 'TEA', 'rabbita_codemirror', 'rabbita_xterm'.

Standardize delegation from Opus to Sonnet/Haiku agents. Use when about to spawn an Agent() for implementation, debugging, review, or mechanical tasks. Provides routing checklist, handoff format, and task-type templates.

Tracks deprecated MoonBit syntax to avoid generating invalid code. Reference this before writing MoonBit code. Auto-maintained when deprecated patterns are discovered.

A skill for creating and maintaining a user-facing changelog. It combines the automation of analyzing git commits with the clear, human-readable structure of the "Keep a Changelog" standard and suggests the next semantic version number.

Use when writing or reviewing MoonBit code against the `dowdiness/incr` reactive library (v0.14.x+) — building Inputs/Deriveds/ReachableDeriveds, attaching long-lived derived cells with `Watch`/`Observer`, adding microbenchmarks, or wrapping a reactive pipeline in a struct. Catches recurring idiom misses (inside-vs-outside read semantics, GC-anchor `Watch`/`Observer`, `Type::Type` constructor naming, defensive copies).

Use when writing or reviewing MoonBit code against the `dowdiness/loom` framework — constructing parsers with `new_parser`, updating `Parser` via `apply_edit`/`set_source`, attaching downstream `@incr.Derived` pipelines to `parser.runtime()`, or preserving semantic projection IDs with `ProjectionIdentityBaseline` / `ProjectionIdentityTracker`. Catches recurring mistakes such as constructing `@incremental.ImperativeParser` inside a reactive closure or advancing projection identity baselines before semantic lowering succeeds.

Use this skill when the user asks to save, remember, recall, or organize memories. Triggers on: 'remember this', 'save this', 'note this', 'what did we discuss about...', 'check your notes', 'clean up memories'. Also use proactively when discovering valuable findings worth preserving.