Plugin Claude Code
Agent skills from Nacif — production workflows and tooling, shared openly with the community.
2 1mo ago A
tokens not measured
original MIT
Instructions file
Instructions for nacif-software/skills, covering claude.md, what this repo is, where the authoritative guidance lives, validating a skill and the two skill types.
2 1mo ago A 1,098 tokens
original MIT
Skill Claude CodeCodex
Use when a repo has a backlog of open Dependabot PRs and the user wants them consolidated into a few CI-validated aggregate PRs — not reviewed individually and not auto-merged. Triggers on "aggregate dependabot", "bundle dependency PRs", "consolidate dependabot PRs", or "clean up the dependabot backlog".
2 1mo ago A 77 tokens
original MIT
Skill Claude CodeCodex
Use when delegating work to subagents, handing work to another agent, splitting implementation tasks, or trying to reduce prompt/context bloat.
2 1mo ago A 33 tokens
original MIT
ddd-planning
05Skill Claude CodeCodex
DDD-guided feature planning — walk strategic alignment, model, tactical and supple design before building. Use when planning a feature through Domain-Driven Design, defining bounded contexts, aggregates, or the ubiquitous language.
2 1mo ago A 45 tokens
original MIT
ddd-review
06Skill Claude CodeCodex
Review code in the light of Domain-Driven Design. Invoke on a diff (default) or a file/directory. Use when checking that an implementation faithfully reflects the domain model — ubiquitous language, layer isolation, aggregates, supple design.
2 1mo ago A 50 tokens
original MIT
Skill Claude CodeCodex
Use when the user wants to design a feature, turn an idea into a spec, write a design doc, create an implementation plan, or optionally continue into implementation.
2 1mo ago A 36 tokens
original MIT
Skill Claude CodeCodex
TEMPLATE — not a working skill. Copy this folder as the starting point for a real skill, then rewrite everything. A good description names what the skill does and the concrete triggers, e.g. "Use when the user asks to draft a release changelog from merged PRs. Triggers on mentions of changelog, release notes, or 'what…
2 1mo ago A 82 tokens
original MIT
execute-plan
09Skill Claude CodeCodex
Use when an approved or reviewed implementation plan, issue breakdown, or task list is ready to be implemented by one or more agents.
2 1mo ago A 29 tokens
original MIT
Skill Claude CodeCodex
Implements tickets or items one at a time using /implement serially with subagents. Use when the user provides a batch/list of things to implement.
2 1mo ago A 34 tokens
original MIT
Skill Claude CodeCodex
Use when a design doc, PRD, issue, or approved requirements need to become a concrete implementation plan before coding or delegating work.
2 1mo ago A 32 tokens
original MIT
Skill Claude CodeCodex
Use when the user wants to write a changelog or release notes for a production deploy and (optionally) post them to a team chat channel. Triggers on mentions of changelog, release notes, deploy notes, prod deploy announcement, "post the changelog", "changelog for PR ", or "what's shipping in this deploy".
2 1mo ago A 77 tokens
original MIT
pr-boundary
13Skill Claude CodeCodex
Use when deciding whether work should be one PR, stacked PRs, separate PRs, or follow-up work; triggers on PR scope, PR split, stacked PR, branch too large, reviewability, rollout, rollback, or separating refactors from feature work.
2 1mo ago A 57 tokens
original MIT
pr-wrap-up
14Skill Claude CodeCodex
Use when implementation on a PR branch is done and the PR needs to be finalized for human review, without an external second-model review loop. Triggers on "wrap up the PR", "finish the PR", "get this ready for review", or a request to clean up and ship a draft PR.
2 1mo ago A 65 tokens
original MIT
Skill Claude CodeCodex
Rewrite the title and description of the current branch's open PR from its diff, blind to the current title and body. Triggers on "rewrite the PR description", "write the PR title and description", or "describe this PR". Never opens a PR.
2 1mo ago A 57 tokens
original MIT
Skill Claude CodeCodex
Use when a PR branch is done and the user wants it hardened by an external Codex review loop (via the official OpenAI Codex plugin for Claude Code) before being finalized for human review. Triggers on "review and wrap up the PR", "codex review then ship", "harden and finish the PR", "run a codex review loop on this…
2 1mo ago A 83 tokens
original MIT
Skill Claude CodeCodex
Use when the user wants to review open Dependabot PRs in a repository, check dependency changelogs for breaking changes, and get a merge-safety verdict for each. Triggers on mentions of Dependabot, dependency update PRs, "review the dependency PRs", "are these bumps safe to merge", or batch-reviewing open dependency…
2 1mo ago A 79 tokens
original MIT
review-plan
18Skill Claude CodeCodex
Use when an implementation plan, task breakdown, issue sequence, test strategy, PR scope, or subagent split needs review before coding, delegation, or execution.
2 1mo ago A 35 tokens
original MIT
review-pr
19Skill Claude CodeCodex
Use when the user asks to review a pull request, review a branch or diff, comment review feedback, apply PR review changes, or choose between review-only and fix mode.
2 1mo ago A 38 tokens
original MIT
scope-check
20Skill Claude CodeCodex
Reviews a feature implementation or plan against the original ask — flags scope beyond the minimum version, complexity beyond the simplest implementation, and at most three strictly-capped low-hanging-fruit wins, reporting findings without applying changes. Use when a feature implementation or plan exists (working…
2 1mo ago A 121 tokens
original MIT
Skill Claude CodeCodex
Use after implementation or before PR readiness when comparing a diff, branch, or completed task against a design doc, implementation plan, PR boundary, test strategy, issue, acceptance criteria, or promised scope.
2 1mo ago A 45 tokens
original MIT
Skill Claude CodeCodex
Use when execute-plan or another coordinator needs to run one plan task, or a batch of independent tasks, through fresh-subagent implementation and spec review: dispatch an implementer, verify it against the task's acceptance criteria, route fixes back to the same implementer, and mark the task done once spec review…
2 1mo ago A 101 tokens
original MIT
Skill Claude CodeCodex
Use when planning, reviewing, or improving tests for a feature, bugfix, PR, implementation plan, test matrix, verification plan, regression risk, edge cases, manual checks, or deciding what not to test.
2 1mo ago A 47 tokens
original MIT
Skill Claude CodeCodex
Use when claiming work is complete, tests pass, a bug is fixed, a review finding is addressed, or before handing off, committing, pushing, or opening a PR.
2 1mo ago A 39 tokens
original MIT