All authors
Claude Skills by CodeForBreakfast
github.com/CodeForBreakfast8 skills0 installs14 views
- Effect Model AuditUse when asked to audit the type modelling of beady-eye's Claude Code plugin — whether its domain types are honest, whether they can represent illegal states, or for a periodic model-level type review. Triggers on "audit our type modelling", "are our types honest", "can our types represent illegal states", "model-level type review", "audit our domain types". The judgement sibling to effect-native-audit: a workflow-driven reasoning sweep over the plugin's domain types, not an anchored regressi...Votes: 0GitHub stars: 15
- Effect Native AuditUse when asked to audit the Claude Code plugin under plugin/ for places it hand-rolled what Effect ships natively, or to run the Effect-native sweep. Triggers on "effect-native audit", "where are we reinventing Effect", "rolled our own instead of an Effect builtin", "sweep for native Effect helpers", "audit our FP/Effect modelling", "audit our concurrency/error handling". A workflow-driven substitution, structural, modelling and behaviour audit — not a substitute for the Effect language servi...Votes: 0GitHub stars: 15
- RenovateUse when asked to triage, review, evaluate, drain or process beady-eye's Renovate pull requests — the minor, major and lock-file-maintenance bumps Renovate opens and holds rather than merging. Triggers on "review the renovate PRs", "triage renovate", "drain the dependency queue", "the held bumps". Not for cutting a release, and not for the patch and digest updates Renovate lands on its own.Votes: 0GitHub stars: 15
- Using Beady EyeUse when a session waits on a bead it cannot move itself, such as a question it asked, another agent's work or a pull request merging, and when a `<channel source="beady-eye">` block arrives. Covers what to watch, how to read a change, what to do when the watcher is down, and unwatching a bead once the session is done waiting on it.Votes: 0GitHub stars: 15
- Effect Model AuditUse when asked to audit this codebase's type modelling — whether the domain types are honest, whether they can represent illegal states, or for a periodic model-level type review. Triggers on "audit our type modelling", "are our types honest", "can our types represent illegal states", "model-level type review", "audit our domain types". The JUDGMENT sibling to effect-native-audit — a workflow-driven reasoning sweep over the port/domain type surface, not the anchor-driven regression net and no...Votes: 0GitHub stars: 2
- Effect Native AuditUse when asked to audit this codebase for places we hand-rolled what Effect ships natively, or to run the Effect-native sweep. Triggers on "effect-native audit", "where are we reinventing Effect", "rolled our own instead of an Effect builtin", "sweep for native Effect helpers", "audit our FP/Effect modelling", "audit our concurrency/error handling". A workflow-driven structural, modelling & behaviour audit — not a substitute for the @effect/language-service idiom linter.Votes: 0GitHub stars: 2
- RenovateUse when explicitly asked to triage, review, evaluate, drain, or process Renovate dependency PRs — the held minor/major bumps Renovate opens but does not auto-merge. Triggers on phrases like "review the renovate PRs", "triage renovate", "drain the dependency queue", "evaluate the held bumps". Not for cutting a release or for the patch updates Renovate already auto-merges.Votes: 0GitHub stars: 2
- Using CommyUse when communicating on commy — choosing a channel, naming a topic, when to be terse vs substantive, when to react instead of reply, when to mention a human. Triggers any time you're about to post, subscribe, or are unsure where a message belongs.Votes: 0GitHub stars: 2