From ironforge
Implements Sudocode issues one by one in dependency order using strict TDD (RED → GREEN → REFACTOR → COMMIT). Activates Context7 for library documentation and leverages LSP diagnostics when available. Updates issue status in Sudocode after each commit.
npx claudepluginhub romaindecoster/ironforge --plugin ironforgeThis skill is limited to using the following tools:
You are executing the Ironforge implementation phase. Work through Sudocode issues in dependency
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
You are executing the Ironforge implementation phase. Work through Sudocode issues in dependency order, one at a time. Follow the TDD cycle strictly for every issue.
$ARGUMENTS
/ironforge:review..sudocode/ to get the full list of specs and issuesIf a blocker is detected (dependency issue not yet complete), stop and report:
Issue [id] is blocked by [id] — implement that one first, or run
/ironforge:implement [id]
If graphify-out/ exists in the project root, extract the target function or module name
from the issue title or description, then invoke ironforge:graph-explore:
Skill({ skill: "ironforge:graph-explore", args: "Qui appelle {cible} ? Que fait-elle appeler ? Blast radius ?" })
Replace {cible} with the actual function/module name from the issue.
If the skill returns content, use it to extend the list of files to read in Step 2 —
callers and callees may need to be read even if not listed in the issue.
If graphify-out/ is absent or ironforge:graph-explore returns empty, proceed directly to Step 2.
Before writing any code:
Activate @bmad-agent-dev. Follow this cycle strictly. Do not skip or merge steps.
# Rust
cargo test [test_name] 2>&1
# TypeScript / Node
npm test -- --testNamePattern="[test_name]" 2>&1
# Python
pytest [test_file]::[test_name] -v 2>&1
use context7— fetch documentation for [library@version]
cargo test 2>&1 | tail -5
# or npm test / pytest equivalent
git add [files changed]
git commit -m "[issue-id] [short description of what was implemented]"
Commit message format: i-xxxx Short description in imperative mood
After the commit, mark the issue as complete via upsert_issue():
doneCheck the remaining issue list:
## Implementation — Done
All issues implemented and committed.
**Next step:** /ironforge:review