소스 정보
- 저장소
- benbrastmckie/nvim
- 최근 소스 활동
- 2026년 8월 10일 06:35
- 감지된 SKILL.md 언어
- 영어
- 스타
- 445
- 포크
- 461
설치 방법
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
소스 파일 검토
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
메뉴
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
SOC 직업 분류 기준
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/benbrastmckie/nvim --skill skill-consult명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SKILL.md 표시 중
Orchestrate multi-agent implementation with parallel phase execution. Spawns teammates for independent phases and coordinates dependent phases. Includes debugger teammate for error recovery.
Implement CSLib proofs with hard-mode contracts (H2 anti-analysis, H7 territory, H9 wrap-up with sorry_inventory). Invoke for --hard cslib implementation tasks.
Research CSLib formalization patterns with hard-mode contracts (H2 anti-analysis, H3 reference grounding with BibKey verification, H4 adversarial verification). Invoke for --hard cslib research tasks.
| name | skill-consult |
| description | Route design consultations to domain-specific design partner agents |
| allowed-tools | Agent, Bash, Edit, Read, Write |
Thin wrapper that routes design consultation requests to domain-specific design partner agents.
IMPORTANT: This skill implements the skill-internal postflight pattern. After the subagent returns, this skill handles all postflight operations (artifact linking, git commit) before returning.
Reference (do not load eagerly):
.claude/context/formats/subagent-return.mdNote: This skill is a thin wrapper. Context is loaded by the delegated agent, not this skill.
This skill activates when:
/consult --legal commanddomain=legal routes to legal-analysis-agentdomain=investor, domain=technical, domain=competitorDo not invoke for:
Validate required inputs:
domain - Required, currently only "legal" supportedinput_type - Required, one of: file_path, inline_text, design_question, task_numbertask_number - Required (always provided by consult.md Gate In)session_id - Required# Validate domain
case "$domain" in
legal) agent="legal-analysis-agent" ;;
*) return error "Unsupported domain: $domain. Currently supported: legal" ;;
esac
# Validate input type
case "$input_type" in
file_path|inline_text|design_question|task_number) ;;
*) return error "Invalid input_type: $input_type" ;;
esac
Resolve task directory for artifact storage. task_number is always provided by the command's Gate In:
task_data=$(jq -r --argjson num "$task_number" \
'.active_projects[] | select(.project_number == $num)' \
specs/state.json)
if [ -z "$task_data" ]; then
return error "Task $task_number not found"
fi
project_name=$(echo "$task_data" | jq -r '.project_name')
padded_num=$(printf "%03d" "$task_number")
task_dir="specs/${padded_num}_${project_name}"
metadata_file="${task_dir}/.return-meta.json"
mkdir -p "$task_dir"
{
"task_context": {
"description": "{description or input summary}",
"task_type": "founder"
},
"domain": "legal",
"input_type": "{file_path | inline_text | design_question | task_number}",
"file_path": "{file path if applicable}",
"inline_text": "{inline text if applicable}",
"design_question": "{design question if applicable}",
"task_number": "{task_number}",
"metadata_file_path": "{metadata_file}",
"metadata": {
"session_id": "{session_id}",
"delegation_depth": 1,
"delegation_path": ["orchestrator", "consult", "skill-consult"]
}
}
CRITICAL: You MUST use the Agent tool to spawn the agent.
Required Tool Invocation:
Tool: Agent (NOT Skill, NOT Plan)
Parameters:
- subagent_type: "{agent}" (e.g., "legal-analysis-agent")
- prompt: [Include all delegation context fields]
- description: "Legal design consultation"
The agent will:
CRITICAL: If you performed the work above WITHOUT using the Agent tool (i.e., you read files,
wrote artifacts, or updated metadata directly instead of spawning a subagent), you MUST write a
.return-meta.json file now before proceeding to postflight. Use the schema from
return-metadata-file.md with the appropriate status value for this operation.
If you DID use the Agent tool, skip this stage -- the subagent already wrote the metadata.
The following stages MUST execute after work is complete, whether the work was done by a subagent or inline (Stage 4b). Do NOT skip these stages for any reason.
if [ -f "$metadata_file" ] && jq empty "$metadata_file" 2>/dev/null; then
status=$(jq -r '.status' "$metadata_file")
artifact_path=$(jq -r '.artifacts[0].path // ""' "$metadata_file")
artifact_type=$(jq -r '.artifacts[0].type // ""' "$metadata_file")
artifact_summary=$(jq -r '.artifacts[0].summary // ""' "$metadata_file")
else
status="failed"
fi
Link the consultation artifact to the task in state.json:
if [ -n "$artifact_path" ]; then
# Add consultation artifact to state.json
bash .claude/scripts/state-write.sh \
'(.active_projects[] | select(.project_number == $num)).artifacts += [{"path": $path, "type": $type, "summary": $summary}]' \
--session-id "$session_id" \
--arg path "$artifact_path" \
--arg type "$artifact_type" \
--arg summary "$artifact_summary" \
--argjson num "$task_number"
fi
Apply targeted staging per .claude/context/standards/git-staging-scope.md — never a repo-wide
add:
git add "${task_dir}/" "specs/TODO.md" "specs/state.json"
git commit -m "task ${task_number}: legal design consultation
Session: ${session_id}
"
Design consultation complete ({domain}):
- Input: {input type and description}
- Translation gaps: {N} identified
- Consultation report: {artifact_path}
- Task: #{task_number}
- Advisory: recommend attorney review for high-stakes materials
Brief text summary (NOT JSON).
Return immediately with supported domain list.
Return error, do not proceed.
Return partial status with progress from agent output.
Non-blocking: Log failure but continue.