Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
49bb6a8
docs: add Secure Agent Design production guide
cursoragent Aug 4, 2026
7877aa1
docs: tighten Secure Agent Design against codebase realities
cursoragent Aug 4, 2026
d148131
docs: fix Secure Agent Design examples to valid CrewAI APIs
cursoragent Aug 4, 2026
5a158e6
docs: simplify Secure Agent Design examples and fix edge links
cursoragent Aug 4, 2026
481caf0
docs: drop production checklist from Secure Agent Design
cursoragent Aug 4, 2026
acd6eff
Merge branch 'main' into cursor/secure-agent-design-612d
theCyberTech Aug 4, 2026
4d85870
docs: clarify single-agent kickoff control differences
cursoragent Aug 4, 2026
e612ac2
Merge branch 'main' into cursor/secure-agent-design-612d
theCyberTech Aug 7, 2026
112683b
Merge branch 'main' into cursor/secure-agent-design-612d
theCyberTech Aug 9, 2026
da28681
docs: correct kickoff path to AgentExecutor
cursoragent Aug 12, 2026
fbd25df
docs: rewrite Secure Agent Design in plain reference style
cursoragent Aug 12, 2026
23704da
docs: fix locale links to EN-only A2A and E2B pages
cursoragent Aug 12, 2026
ed030e4
ci: pass --use-system-ca to paths-filter under Node 24
cursoragent Aug 12, 2026
13e0f2b
Revert "ci: pass --use-system-ca to paths-filter under Node 24"
cursoragent Aug 12, 2026
3348df0
docs: rewrite Secure Agent Design in simplified technical English
cursoragent Aug 17, 2026
7330582
Merge branch 'main' into cursor/secure-agent-design-612d
theCyberTech Aug 17, 2026
18ceda9
Merge branch 'main' into cursor/secure-agent-design-612d
theCyberTech Aug 24, 2026
ff7fd10
docs: tighten Secure Agent Design scope and fill examples
cursoragent Aug 24, 2026
79f01b4
docs: treat HITL as approval, not a control
cursoragent Aug 24, 2026
d3e3a2c
docs: drop repeated HITL warnings
cursoragent Aug 24, 2026
102bd51
docs: pass validated structured notes in isolation example
cursoragent Aug 24, 2026
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 8 additions & 4 deletions docs/docs.json
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,8 @@
"group": "Agents",
"icon": "user",
"pages": [
"edge/en/guides/agents/crafting-effective-agents"
"edge/en/guides/agents/crafting-effective-agents",
"edge/en/guides/agents/secure-agent-design"
]
},
{
Expand Down Expand Up @@ -13361,7 +13362,8 @@
"group": "Agentes",
"icon": "user",
"pages": [
"edge/pt-BR/guides/agents/crafting-effective-agents"
"edge/pt-BR/guides/agents/crafting-effective-agents",
"edge/pt-BR/guides/agents/secure-agent-design"
]
},
{
Expand Down Expand Up @@ -25711,7 +25713,8 @@
"group": "에이전트 (Agents)",
"icon": "user",
"pages": [
"edge/ko/guides/agents/crafting-effective-agents"
"edge/ko/guides/agents/crafting-effective-agents",
"edge/ko/guides/agents/secure-agent-design"
]
},
{
Expand Down Expand Up @@ -38481,7 +38484,8 @@
"group": "الوكلاء",
"icon": "user",
"pages": [
"edge/ar/guides/agents/crafting-effective-agents"
"edge/ar/guides/agents/crafting-effective-agents",
"edge/ar/guides/agents/secure-agent-design"
]
},
{
Expand Down
6 changes: 6 additions & 0 deletions docs/edge/ar/concepts/agents.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -350,6 +350,12 @@ result = researcher.kickoff("What are the latest developments in language models
print(result.raw)
```

<Note>
يشغّل `kickoff()` مُنفّذ `AgentExecutor`. ولا ينشئ Task ولا Crew.
يحتفظ الـ Agent بدوره وهدفه وخلفيته وأدواته. تُرجع الدالة
`LiteAgentOutput`.
</Note>

## اعتبارات مهمة وأفضل الممارسات

### الأمان وتنفيذ الكود
Expand Down
5 changes: 5 additions & 0 deletions docs/edge/ar/concepts/production-architecture.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -154,9 +154,14 @@ flow.kickoff(restore_from_state_id="<previous-run-state-id>")

يحصل التشغيل الجديد على `state.id` جديد (مولّد تلقائيًا، أو `inputs["id"]` إذا تم تثبيته) لذا لا تمتد كتابات `@persist` الخاصة به إلى تاريخ المصدر. الجمع مع `from_checkpoint` يطلق `ValueError`؛ اختر مصدر ترطيب واحدًا.

## الأمان

يمكن للـ Agents المزودة بأدوات تنفيذ إجراءات حقيقية. راجع [تصميم Agent الآمن](/edge/ar/guides/agents/secure-agent-design) للحد من هذا الخطر.

## الخلاصة

- **ابدأ بتدفق.**
- **حدد حالة واضحة.**
- **استخدم الأطقم للمهام المعقدة.**
- **انشر مع API واستمرارية.**
- اتبع [تصميم Agent الآمن](/edge/ar/guides/agents/secure-agent-design).
2 changes: 2 additions & 0 deletions docs/edge/ar/guides/agents/crafting-effective-agents.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,8 @@ mode: "wide"

سيساعدك هذا الدليل على إتقان فن تصميم الـ Agent، مما يمكّنك من إنشاء شخصيات AI متخصصة تتعاون بفعالية وتفكر بشكل نقدي وتنتج مخرجات عالية الجودة مصممة لاحتياجاتك المحددة.

إذا كانت الـ Agents تستخدم أدوات أو محتوى غير موثوق، فاقرأ أيضًا [تصميم Agent الآمن](/edge/ar/guides/agents/secure-agent-design).

### لماذا يهم تصميم الـ Agent

الطريقة التي تعرّف بها الـ Agents تؤثر بشكل كبير على:
Expand Down
356 changes: 356 additions & 0 deletions docs/edge/ar/guides/agents/secure-agent-design.mdx

Large diffs are not rendered by default.

2 changes: 2 additions & 0 deletions docs/edge/ar/mcp/security.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -147,3 +147,5 @@ mode: "wide"
من خلال فهم اعتبارات الأمان هذه وتنفيذ أفضل الممارسات، يمكنك الاستفادة بأمان من قوة خوادم MCP في مشاريع CrewAI.
هذه ليست شاملة بأي حال، لكنها تغطي المخاوف الأمنية الأكثر شيوعاً وأهمية.
ستستمر التهديدات في التطور، لذا من المهم البقاء على اطلاع وتكييف إجراءات الأمان وفقاً لذلك.

راجع أيضًا [تصميم Agent الآمن](/edge/ar/guides/agents/secure-agent-design).
6 changes: 3 additions & 3 deletions docs/edge/en/concepts/agents.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -638,9 +638,9 @@ asyncio.run(main())
```

<Note>
The `kickoff()` method uses a `LiteAgent` internally, which provides a simpler
execution flow while preserving all of the agent's configuration (role, goal,
backstory, tools, etc.).
`kickoff()` runs an `AgentExecutor`. It does not create a Task or a Crew.
The agent keeps its role, goal, backstory, and tools. The method returns
`LiteAgentOutput`.
</Note>

## Important Considerations and Best Practices
Expand Down
5 changes: 5 additions & 0 deletions docs/edge/en/concepts/production-architecture.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -154,9 +154,14 @@ flow.kickoff(restore_from_state_id="<previous-run-state-id>")

The new run gets a fresh `state.id` (auto-generated, or `inputs["id"]` if pinned) so its `@persist` writes don't extend the source's history. Combining with `from_checkpoint` raises a `ValueError`; pick one hydration source.

## Security

Agents with tools can take real-world actions. See [Secure Agent Design](/edge/en/guides/agents/secure-agent-design) to limit that risk.

## Summary

- **Start with a Flow.**
- **Define a clear State.**
- **Use Crews for complex tasks.**
- **Deploy with an API and persistence.**
- Follow [Secure Agent Design](/edge/en/guides/agents/secure-agent-design).
4 changes: 4 additions & 0 deletions docs/edge/en/guides/agents/crafting-effective-agents.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,10 @@ At the heart of CrewAI lies the agent - a specialized AI entity designed to perf

This guide will help you master the art of agent design, enabling you to create specialized AI personas that collaborate effectively, think critically, and produce high-quality outputs tailored to your specific needs.

<Tip>
If agents use tools or untrusted content, also read [Secure Agent Design](/edge/en/guides/agents/secure-agent-design).
</Tip>

### Why Agent Design Matters

The way you define your agents significantly impacts:
Expand Down
Loading
Loading