メニュー
Use when user wants to search for existing knowledge, recall a specific learning, or discover what knowledge is available.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/joenandez/spectre-labs --skill spectre-recallコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
Use this skill when you see @name or /name patterns in user input without file identifiers. This means the user is trying to call a subagent or slash command using Spectre Agent Tools. The @ prefix dispatches to a subagent (e.g., '@tdd-agent write tests', '@code-reviewer check this file'). The / prefix executes a slash command (e.g., '/spectre:scope', '/spectre:plan'). Both require Spectre CLI to run specialized agents and predefined prompts.
Use when modifying build loop code, debugging stats/token tracking, adding CLI features, changing iteration prompts, or understanding how spectre-build works end-to-end
Use when modifying the plan loop, debugging plan stages, changing clarification flow, scope isolation, or understanding how spectre-build --plan works end-to-end
| name | spectre-recall |
| description | Use when user wants to search for existing knowledge, recall a specific learning, or discover what knowledge is available. |
Search and load relevant knowledge from the project's spectre learnings into your context.
# SPECTRE Knowledge Registry
# Format: skill-name|category|triggers|description
feature-build-loop|feature|build loop, spectre-build, build iteration, validation cycle, promise tags, build stats, build dashboard, token tracking, code review, phase awareness, git scope, hooks|Use when modifying build loop code, debugging stats/token tracking, adding CLI features, changing iteration prompts, or understanding how spectre-build works end-to-end
feature-plan-pipeline|feature|plan pipeline, --plan, --scope-name, planning loop, plan stages, clarifications, plan resume, scope to manifest, scope isolation, scope slug, run_plan_pipeline, create_plan_pipeline|Use when modifying the planning pipeline, debugging plan stages, changing clarification flow, scope isolation, or understanding how spectre-build --plan works end-to-end
feature-ship-pipeline|feature|ship pipeline, --ship, ship loop, clean stage, test stage, rebase stage, land branch, run_ship_pipeline, create_ship_pipeline, ship hooks, notify_ship_complete, clean_discover, clean_investigate, clean_execute, test_plan, test_execute, test_verify, test_commit, sub-stage, subagent dispatch|Use when modifying the ship pipeline, debugging ship stages, changing clean/test/rebase behavior, or understanding how spectre-build --ship works end-to-end
strategy-scaling-architecture|strategy|scale, GUI, multi-model, model abstraction, event bus, orchestrator, async, adversarial review, live steering, telemetry, industrial, node editor, pipeline editor, multi-tenant, server layer, product scaling|Use when planning product scaling, adding GUI/server layers, multi-model support, adversarial reviews, live steering, telemetry, or node-based pipeline editors to spectre-build
Skill({skill-name})/recall {query} — search registry for matches/recall — show all available knowledge by categorySingle match → Load automatically via Skill({skill-name})
Multiple matches → List options, ask user which to load
No matches → Suggest /learn to capture new knowledge