research
Delegate Broad Codebase Research
Large investigations can flood the primary agent context with irrelevant files and logs. This skill delegates read-only research and returns concise conclusions with cited evidence.
Установить с помощью моего Агента
Скопируйте этот запрос в своего Агента. Он содержит каноническую страницу Skill и манифест.
Review the Skillstore skill "research" from https://skillstore.io/skills/warpdotdev-research.md and its manifest at https://skillstore.io/api/skills/warpdotdev-research/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Ваш Агент по-прежнему должен показать план и запросить все подтверждения, требуемые политикой безопасности.
Ресурсы для AI-агентов
Используйте эти ссылки, когда AI-агенту, crawler или script нужен чистый контекст вместо полной страницы.
Протестировать
Использование «research». Find the root cause of this failing integration test and cite the relevant implementation and test locations.
Ожидаемый результат:
- Root cause: The retry handler converts the timeout into a permanent failure.
- Evidence: The report identifies the handler and failing test locations.
- Caveat: External service behavior was not reproduced locally.
Использование «research». Summarize this large pull request, including behavior changes, risks, and test coverage.
Ожидаемый результат:
- Change: The request pipeline now validates credentials before loading account data.
- Risk: Existing clients with incomplete credentials may fail earlier.
- Coverage: Unit tests cover validation, but migration behavior remains untested.
Использование «research». Explain how session recovery works across the repository and identify the main call sites.
Ожидаемый результат:
- Flow: Recovery starts after connection loss, restores persisted state, and reconnects active sessions.
- Evidence: The summary lists the coordinator, persistence layer, and reconnect entry points.
- Unknown: Concurrent recovery behavior requires runtime verification.
Аудит безопасности
БезопасноAll five static findings are false positives. Four are Markdown inline-code spans, while line 36 discusses subagent costs without performing reconnaissance. No prompt injection, command execution, or semantic security issue was found.
Факторы риска
⚙️ Внешние команды (4)
Поделиться и цитировать этот отчет
Делитесь версионным отчетом об оценке, нейтральным значком, встраиваемой карточкой и цитатами. Skillstore публикует доказательства, не решая, безопасен ли этот Skill.
Копировать ссылку на отчёт
https://skillstore.io/skills/warpdotdev-research/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportЗначок Markdown
[](https://skillstore.io/skills/warpdotdev-research?utm_source=security_passport_badge)Значок HTML
<a href="https://skillstore.io/skills/warpdotdev-research?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/warpdotdev-research/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Встраиваемая карточка
<iframe src="https://skillstore.io/embed/skills/warpdotdev-research.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Академические ссылки (APA · BibTeX · CFF)
Цитата APA
warpdotdev. (2026). research security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/warpdotdev-research/audits/1Цитата BibTeX
@techreport{warpdotdev-warpdotdev-research-2026,
author = {warpdotdev},
title = {research security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/warpdotdev-research/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "research security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "warpdotdev"
date-released: "2026-09-04"
url: "https://skillstore.io/skills/warpdotdev-research/audits/1"
identifiers:
- type: other
value: "skillstore:warpdotdev-research:audit:1"
description: "Skillstore immutable audit report identifier"
Сравнить варианты
2 устанавливаемых вариантовНавык каждого автора остаётся отдельным устанавливаемым вариантом. Рекомендуемый вариант ранжируется по данным Skillstore.
Почему этот вариант первый
mattpocock-research
2026-09-09
warpdotdev-research
2026-09-09
Оценка Skillstore
Почему такая оценка Достоверность доказательств: СреднийЧто вы можете построить
Trace a complex test failure
Delegate inspection of long CI logs and related code, then receive the likely cause with supporting locations.
Summarize a large pull request
Review a substantial diff through a subagent and receive the main changes, rationale, risks, and evidence.
Survey an unfamiliar architecture
Map how a component or API is used across a repository without loading every inspected file into primary context.
Попробуйте эти промпты
Investigate {question} in {repository_path}. Work read-only. Return a concise answer with exact file paths, symbols, and important caveats.Analyze {log_or_test_failure} and trace the relevant code. Identify the root cause, cite supporting locations, and note anything not verified.Research {pull_request_or_diff}. Summarize behavioral changes, affected components, design rationale, risks, and tests. Support each major conclusion with file locations.Investigate these independent areas in parallel: {area_one}, {area_two}, and {area_three}. Synthesize the results, preserve evidence, and identify cross-area uncertainties.Лучшие практики
- Provide an exact question, repository path, and suspected files or symbols.
- Keep investigations read-only unless the task explicitly requires changes.
- Request concise conclusions, exact evidence locations, caveats, and unresolved questions.
Избегать
- Do not delegate a single lookup or a few known file reads.
- Do not delegate when the primary agent must edit the same files immediately.
- Do not split one coherent investigation into artificial parallel tasks.
Часто задаваемые вопросы
When should I use this skill?
When should I inspect files directly?
Does the skill modify repository files?
Can it run several investigations at once?
What should a research report include?
Which AI tools support this skill?
Сведения для разработчиков
Автор
warpdotdevЛицензия
MIT
Ревизия Skillstore
r1
Примечание о версии
Автор не указал версию.
Ссылка
e5464e662405de6361fdde6b984f9ea865635f64
Актуальность поддержки
08.09.2026
Использование
1 загрузок · 0 просмотров
Структура файлов
📄 SKILL.md