From xlfg-engineering
Internal xlfg phase skill. Use only during /xlfg runs to gather repo truth, current constraints, harness facts, and targeted external research when needed.
npx claudepluginhub flrngel/xlfg --plugin xlfg-engineeringThis skill is limited to using the following tools:
Use only during `/xlfg` orchestration.
Researches implementation approaches for a project phase by validating roadmap, checking prior research, gathering context, and spawning gsd-phase-researcher agent. Standalone for feasibility or pre-planning.
Researches implementation approaches for a specific project phase from ROADMAP.md by normalizing input, checking existing research, gathering context, and spawning a kata-phase-researcher agent.
Auto-routes pre-implementation context gathering: detects codebase vs external research needs and dispatches workflows. Triggers on 'get context', 'how does X work', architecture queries, or setup-before-coding requests.
Share bugs, ideas, or general feedback.
Use only during /xlfg orchestration.
Input: $ARGUMENTS (RUN_ID or latest)
Gather the repo and product truth needed for an honest plan without exploding context or duplicating state.
RUN_ID, DOCS_RUN_DIR, and DX_RUN_DIR.context.mdmemory-recall.mdspec.mdworkboard.mddocs/xlfg/knowledge/current-state.mdspec.md. Keep repo exploration scoped to those direct asks, implied asks, objective groups, and blockers.xlfg-repo-mapperxlfg-harness-profiler for any build / bugfix / delivery runxlfg-context-adjacent-investigator, xlfg-context-constraints-investigator, or xlfg-context-unknowns-investigator one at a time when the request is bundled, risky, or still assumption-heavyxlfg-env-doctor when local server behavior is relevantxlfg-researcher only when freshness or missing domain knowledge makes repo truth insufficientStatus:, and contains real findings instead of preparation notes. If it does not, use SendMessage with the returned agent ID to resume the same specialist once before treating the lane as failed. If no agent ID is available, re-dispatch the exact same packet once.context.md with:
spec.md.workboard.md and either retry once or continue only with an explicit gap note.research.md only when external research materially changes the decision surface.workboard.md current while planning is in progress.PRIMARY_ARTIFACT with Status: IN_PROGRESS, the scoped mission, and a short checklist so the specialist is resuming a concrete work item instead of starting from an empty chat turn.PRIMARY_ARTIFACT: <exact path>
FILE_SCOPE: <bounded files or paths>
DONE_CHECK: <single honest check or NONE>
RETURN_CONTRACT: DONE|BLOCKED|FAILED <artifact-path> only
LS or Glob instead of Read on directories; use Grep plus chunked Read windows instead of loading an oversized file in one shot.spec.md.