developer-roadmap changelog

15 days ago

Hey there, tech enthusiasts! 🚀 We've got some exciting updates and improvements to share with you. Our latest changes are all about enhancing your learning journey in the world of AI and cybersecurity. Here’s a sneak peek at what’s new and improved:

New Features

AI Agents Roadmap: Say hello to a brand-new, comprehensive roadmap for AI agents! This roadmap is your go-to guide for diving deep into the world of AI agents, covering everything from tool invocation to safety measures. It's all about providing you with the insights you need to design, build, and ship AI agents by 2025. 📚🤖
AI Red Teaming Roadmap: We've also rolled out a roadmap focused on AI red teaming. It's packed with content on advanced techniques, adversarial examples, and AI security fundamentals. This roadmap is essential for anyone looking to tackle AI system vulnerabilities and threats. 🔍🛡️

Improvements

AI Agents Content Update: We've spruced up the AI Agents roadmap with new content, enhancing your understanding of agent functionalities, tools, and methodologies. Whether you're interested in data privacy or memory management, we've got you covered! 🧠🔧
Cybersecurity Content & Link Adjustments: Our cybersecurity roadmap just got a makeover! We've refactored 40 topics to ensure that the content is crystal clear and the links are spot-on. This update makes exploring cybersecurity topics a breeze. 🛠️🔗
Memory Concepts Clarified: We've updated the descriptions of episodic vs. semantic memory and short-term memory in the AI agents roadmap. These changes highlight practical applications and design considerations, making it easier to grasp these concepts. 🧩📖

Bug Fixes

Syntax Error Handling: We've squashed a syntax error bug in the use-custom-roadmap.ts file. The improved error handling ensures smoother navigation through our custom roadmaps. 🐛✔️
Broken URLs Fixed: We've cleaned up the "Why Red Team AI Systems?" document by removing outdated links, so you can focus on the critical content without distractions. 🌐🔧
Height Adjustment for AI Agents Roadmap: A quick fix to the height specification in the AI Agents roadmap ensures everything looks just right. 📏👌

Chores

Roadmap Content JSON Update: We've updated several JSON files, adding new ones for AI agents and tweaking existing ones for AI data scientists, AI red teaming, and cybersecurity. These changes keep our content fresh and relevant. 🗂️🔄

We hope these updates make your learning journey more enjoyable and insightful. Happy exploring! 🌟

Included Commits

2025-04-25T13:57:24 See commit

The commit titled "AI Red Teaming Roadmap - Initial Commit (#8553)" marks the first submission of a comprehensive roadmap focused on AI red teaming. The commit includes a series of modifications to various markdown files within the project, indicating an organized effort to structure and format content related to advanced techniques, adversarial examples, AI security fundamentals, and numerous other critical topics in AI security and red teaming methodologies.

The changes span multiple aspects of AI security, including but not limited to authentication, data poisoning, model vulnerabilities, and testing methodologies such as black-box and white-box testing. This initial commit lays the groundwork for a detailed framework that aims to enhance understanding and practices in AI red teaming, ensuring that stakeholders are equipped with essential knowledge and strategies to address emerging threats and vulnerabilities in AI systems.

Files changed

src/data/roadmaps/ai-red-teaming/content/advanced-techniques@soC-kcem1ISbnCQMa6BIB.md
src/data/roadmaps/ai-red-teaming/content/adversarial-examples@xjlttOti-_laPRn8a2fVy.md
src/data/roadmaps/ai-red-teaming/content/adversarial-training@2Y0ZO-etpv3XIvunDLu-W.md
src/data/roadmaps/ai-red-teaming/content/agentic-ai-security@FVsKivsJrIb82B0lpPmgw.md
src/data/roadmaps/ai-red-teaming/content/ai-security-fundamentals@R9DQNc0AyAQ2HLpP4HOk6.md
src/data/roadmaps/ai-red-teaming/content/api-protection@Tszl26iNBnQBdBEWOueDA.md
src/data/roadmaps/ai-red-teaming/content/authentication@J7gjlt2MBx7lOkOnfGvPF.md
src/data/roadmaps/ai-red-teaming/content/authentication@JQ3bR8odXJfd-1RCEf3-Q.md
src/data/roadmaps/ai-red-teaming/content/automated-vs-manual@LVdYN9hyCyNPYn2Lz1y9b.md
src/data/roadmaps/ai-red-teaming/content/benchmark-datasets@et1Xrr8ez-fmB0mAq8W_a.md
src/data/roadmaps/ai-red-teaming/content/black-box-testing@0bApnJTt-Z2IUf0X3OCYf.md
src/data/roadmaps/ai-red-teaming/content/code-injection@vhBu5x8INTtqvx6vcYAhE.md
src/data/roadmaps/ai-red-teaming/content/conferences@LuKnmd9nSz9yLbTU_5Yp2.md
src/data/roadmaps/ai-red-teaming/content/confidentiality-integrity-availability@WZkIHZkV2qDYbYF9KBBRi.md
src/data/roadmaps/ai-red-teaming/content/continuous-monitoring@7Km0mFpHguHYPs5UhHTsM.md
src/data/roadmaps/ai-red-teaming/content/continuous-testing@65Lo60JQS5YlvvQ6KevXt.md
src/data/roadmaps/ai-red-teaming/content/countermeasures@G1u_Kq4NeUsGX2qnUTuJU.md
src/data/roadmaps/ai-red-teaming/content/ctf-challenges@2Imb64Px3ZQcBpSQjdc_G.md
src/data/roadmaps/ai-red-teaming/content/custom-testing-scripts@C1zO2xC0AqyV53p2YEPWg.md
src/data/roadmaps/ai-red-teaming/content/data-poisoning@nD0_64ELEeJSN-0aZiR7i.md
src/data/roadmaps/ai-red-teaming/content/direct@5zHow4KZVpfhch5Aabeft.md
src/data/roadmaps/ai-red-teaming/content/emerging-threats@-G8v_CNa8wO_g-46_RFQo.md
src/data/roadmaps/ai-red-teaming/content/ethical-considerations@1gyuEV519LjN-KpROoVwv.md
src/data/roadmaps/ai-red-teaming/content/forums@Smncq-n1OlnLAY27AFQOO.md
src/data/roadmaps/ai-red-teaming/content/generative-models@3XJ-g0KvHP75U18mxCqgw.md
src/data/roadmaps/ai-red-teaming/content/grey-box-testing@ZVNAMCP68XKRXVxF2-hBc.md
src/data/roadmaps/ai-red-teaming/content/indirect@3_gJRtJSdm2iAfkwmcv0e.md
src/data/roadmaps/ai-red-teaming/content/industry-credentials@HHjsFR6wRDqUd66PMDE_7.md
src/data/roadmaps/ai-red-teaming/content/industry-standards@WePO66_4-gNcSdE00WKmw.md
src/data/roadmaps/ai-red-teaming/content/infrastructure-security@nhUKKWyBH80nyKfGT8ErC.md
src/data/roadmaps/ai-red-teaming/content/insecure-deserialization@aKzai0A8J55-OBXTnQih1.md
src/data/roadmaps/ai-red-teaming/content/introduction@HFJIYcI16OMyM77fAw9af.md
src/data/roadmaps/ai-red-teaming/content/jailbreak-techniques@Ds8pqn4y9Npo7z6ubunvc.md
src/data/roadmaps/ai-red-teaming/content/lab-environments@MmwwRK4I9aRH_ha7duPqf.md
src/data/roadmaps/ai-red-teaming/content/large-language-models@8K-wCn2cLc7Vs_V4sC3sE.md
src/data/roadmaps/ai-red-teaming/content/llm-security-testing@xJYTRbPxMn0Xs5ea0Ygn6.md
src/data/roadmaps/ai-red-teaming/content/model-inversion@iE5PcswBHnu_EBFIacib0.md
src/data/roadmaps/ai-red-teaming/content/model-vulnerabilities@uBXrri2bXVsNiM8fIHHOv.md
src/data/roadmaps/ai-red-teaming/content/model-weight-stealing@QFzLx5nc4rCCD8WVc20mo.md
src/data/roadmaps/ai-red-teaming/content/monitoring-solutions@59lkLcoqV4gq7f8Zm0X2p.md
src/data/roadmaps/ai-red-teaming/content/neural-networks@RuKzVhd1nZphCrlW1wZGL.md
src/data/roadmaps/ai-red-teaming/content/prompt-engineering@gx4KaFqKgJX9n9_ZGMqlZ.md
src/data/roadmaps/ai-red-teaming/content/prompt-hacking@1Xr7mxVekeAHzTL7G4eAZ.md
src/data/roadmaps/ai-red-teaming/content/prompt-injection@XOrAPDRhBvde9R-znEipH.md
src/data/roadmaps/ai-red-teaming/content/red-team-simulations@DpYsL0du37n40toH33fIr.md
src/data/roadmaps/ai-red-teaming/content/reinforcement-learning@Xqzc4mOKsVzwaUxLGjHya.md
src/data/roadmaps/ai-red-teaming/content/remote-code-execution@kgDsDlBk8W2aM6LyWpFY8.md
src/data/roadmaps/ai-red-teaming/content/reporting-tools@BLnfNlA0C4yzy1dvifjwx.md
src/data/roadmaps/ai-red-teaming/content/research-groups@ZlR03pM-sqVFZNhD1gMSJ.md
src/data/roadmaps/ai-red-teaming/content/research-opportunities@VmaIHVsCpq2um_0cA33V3.md
src/data/roadmaps/ai-red-teaming/content/responsible-disclosure@KAcCZ3zcv25R6HwzAsfUG.md
src/data/roadmaps/ai-red-teaming/content/risk-management@MupRvk_8Io2Hn7yEvU663.md
src/data/roadmaps/ai-red-teaming/content/robust-model-design@6gEHMhh6BGJI-ZYN27YPW.md
src/data/roadmaps/ai-red-teaming/content/role-of-red-teams@Irkc9DgBfqSn72WaJqXEt.md
src/data/roadmaps/ai-red-teaming/content/safety-filter-bypasses@j7uLLpt8MkZ1rqM7UBPW4.md
src/data/roadmaps/ai-red-teaming/content/specialized-courses@s1xKK8HL5-QGZpcutiuvj.md
src/data/roadmaps/ai-red-teaming/content/supervised-learning@NvOJIv36Utpm7_kOZyr79.md
src/data/roadmaps/ai-red-teaming/content/testing-platforms@c8n8FcYKDOgPLQvV9xF5J.md
src/data/roadmaps/ai-red-teaming/content/threat-modeling@RDOaTBWP3aIJPUp_kcafm.md
src/data/roadmaps/ai-red-teaming/content/unauthorized-access@DQeOavZCoXpF3k_qRDABs.md
src/data/roadmaps/ai-red-teaming/content/unsupervised-learning@ZC0yKsu-CJC-LZKKo2pLD.md
src/data/roadmaps/ai-red-teaming/content/vulnerability-assessment@887lc3tWCRH-sOHSxWgWJ.md
src/data/roadmaps/ai-red-teaming/content/white-box-testing@Mrk_js5UVn4dRDw-Yco3Y.md
src/data/roadmaps/ai-red-teaming/content/why-red-team-ai-systems@fNTb9y3zs1HPYclAmu_Wv.md

2025-04-25T13:59:19 See commit

The commit titled "refactor: (cybersecurity) content & link adjustments" focuses on making significant improvements to 40 topics within the cybersecurity roadmap. This refactoring effort includes modifications to various content files related to cybersecurity, enhancing the clarity and relevance of the material presented. The changes aim to streamline the information and ensure that links and content are accurately aligned, thereby improving the overall user experience for those accessing the roadmap.

The commit highlights a systematic approach to updating resources, as evidenced by the extensive list of modified files, which cover a wide range of cybersecurity topics from operating system hardening to penetration testing and cloud security concepts. This comprehensive update reflects a commitment to maintaining high-quality educational resources in the field of cybersecurity, making it easier for learners and professionals to access up-to-date and relevant information.

Files changed

src/data/roadmaps/cyber-security/content/onedrive@MWqnhDKm9jXvDDjkeVNxm.md
src/data/roadmaps/cyber-security/content/operating-system-hardening@_S25EOGS3P8647zLM5i-g.md
src/data/roadmaps/cyber-security/content/operating-systems@UY6xdt_V3YMkZxZ1hZLvW.md
src/data/roadmaps/cyber-security/content/os-independent-troubleshooting@pJUhQin--BGMuXHPwx3JJ.md
src/data/roadmaps/cyber-security/content/oscp@SwVGVP2bbCFs2uNg9Qtxb.md
src/data/roadmaps/cyber-security/content/packet-sniffers@k6UX0BJho5arjGD2RWPgH.md
src/data/roadmaps/cyber-security/content/parrotos@zR6djXnfTSFVEfvJonQjf.md
src/data/roadmaps/cyber-security/content/penetration-testing-rules-of-engagement@NkAAQikwH-A6vrF8fWpuB.md
src/data/roadmaps/cyber-security/content/perimiter-vs-dmz-vs-segmentation@PUgPgpKio4Npzs86qEXa7.md
src/data/roadmaps/cyber-security/content/picoctf@pou5xHwnz9Zsy5J6lNlKq.md
src/data/roadmaps/cyber-security/content/pki@fxyJxrf3mnFTa3wXk1MCW.md
src/data/roadmaps/cyber-security/content/port-blocking@W7bcydXdwlubXF2PHKOuq.md
src/data/roadmaps/cyber-security/content/port-scanners@iJRQHzh5HXADuWpCouwxv.md
src/data/roadmaps/cyber-security/content/power-shell@paY9x2VJA98FNGBFGRXp2.md
src/data/roadmaps/cyber-security/content/preparation@w6V4JOtXKCMPAkKIQxvMg.md
src/data/roadmaps/cyber-security/content/private-vs-public-keys@7svh9qaaPp0Hz23yinIye.md
src/data/roadmaps/cyber-security/content/private@ecpMKP1cQXXsfKETDUrSf.md
src/data/roadmaps/cyber-security/content/privilege-escalation@cvI8-sxY5i8lpelW9iY_5.md
src/data/roadmaps/cyber-security/content/programming-skills@_RnuQ7952N8GWZfPD60sJ.md
src/data/roadmaps/cyber-security/content/protocol-analyzers@K05mEAsjImyPge0hDtsU0.md
src/data/roadmaps/cyber-security/content/proxmox@jqX1A5hFF3Qznqup4lfiF.md
src/data/roadmaps/cyber-security/content/public@ZDj7KBuyZsKyEMZViMoXW.md
src/data/roadmaps/cyber-security/content/python@XiHvGy--OkPFfJeKA6-LP.md
src/data/roadmaps/cyber-security/content/reconnaissance@UU_inxa8Y2lLP2BRhdLDT.md
src/data/roadmaps/cyber-security/content/replay-attack@mIX8PsIGuwgPCGQZ6ok2H.md
src/data/roadmaps/cyber-security/content/ring@9vEUVJ8NTh0wKyIE6-diY.md
src/data/roadmaps/cyber-security/content/rmf@fjEdufrZAfW4Rl6yDU8Hk.md
src/data/roadmaps/cyber-security/content/rogue-access-point@Ee7LfbhwJbiWjJ3b_bbni.md
src/data/roadmaps/cyber-security/content/roles-of-compliance-and-auditors@kqT0FRLt9Ak9P8PhHldO-.md
src/data/roadmaps/cyber-security/content/route@xFuWk7M-Vctk_xb7bHbWs.md
src/data/roadmaps/cyber-security/content/router@lwSFIbIX-xOZ0QK2sGFb1.md
src/data/roadmaps/cyber-security/content/s3@2jsTgT7k8MeaDtx6RJhOP.md
src/data/roadmaps/cyber-security/content/saas@sVw5KVNxPEatBRKb2ZbS_.md
src/data/roadmaps/cyber-security/content/sans-holiday-hack-challenge@WCeJrvWl837m1BIjuA1Mu.md
src/data/roadmaps/cyber-security/content/shoulder-surfing@FD0bkmxNpPXiUB_NevEUf.md
src/data/roadmaps/cyber-security/content/siem@c2kY3wZVFKZYxMARhLIwO.md
src/data/roadmaps/cyber-security/content/star@P0ZhAXd_H-mTOMr13Ag31.md
src/data/roadmaps/cyber-security/content/switch@r9byGV8XuBPzoqj5ZPf2W.md
src/data/roadmaps/cyber-security/content/true-negative--true-positive@M6uwyD4ibguxytf1od-og.md
src/data/roadmaps/cyber-security/content/tryhackme@kht-L7_v-DbglMYUHuchp.md
src/data/roadmaps/cyber-security/content/understand-the-basics-and-general-flow-of-deploying-in-the-cloud@XL3FVeGFDhAl_gSol6Tjt.md
src/data/roadmaps/cyber-security/content/understand-the-concept-of-infrastructure-as-code@RJctUpvlUJGAdwBNtDSXw.md
src/data/roadmaps/cyber-security/content/understand-the-concept-of-security-in-the-cloud@ThLsXkqLw--uddHz0spCH.md
src/data/roadmaps/cyber-security/content/understand-the-concept-of-serverless@-83ltMEl3le3yD68OFnTM.md
src/data/roadmaps/cyber-security/content/virtualbox@vGVFhZXYOZOy4qFpLLbxp.md
src/data/roadmaps/cyber-security/content/virustotal@rxzcAzHjzIc9lkWSw0fef.md
src/data/roadmaps/cyber-security/content/vmware@AjywuCZdBi9atGUbetlUL.md
src/data/roadmaps/cyber-security/content/vulnhub@W94wY_otBuvVW_-EFlKA6.md
src/data/roadmaps/cyber-security/content/watering-hole-attack@dcvuKHq0nHgHLcLwtl4IJ.md
src/data/roadmaps/cyber-security/content/windows@BNUKEQ4YpZmIhSPQdkHgU.md
src/data/roadmaps/cyber-security/content/winhex@gNan93Mg9Ym2AF3Q2gqoi.md
src/data/roadmaps/cyber-security/content/wireshark@Sm9bxKUElINHND8FdZ5f2.md
src/data/roadmaps/cyber-security/content/wlan@QCVYF1rmPsMVtklBNDNaB.md

2025-04-26T11:18:22 See commit

This commit introduces a new roadmap for AI red teaming, aimed at guiding individuals to become experts in this specialized field. The changes made to the file ai-red-teaming.md include the addition of a brief description and adjustments to the visibility status, changing it from hidden to visible.

The roadmap is designed to provide structured learning and resources for those interested in the nuances of red teaming within the context of artificial intelligence, emphasizing the importance of understanding security vulnerabilities and threat modeling in AI systems. With this update, the roadmap is now accessible to users, encouraging engagement and exploration of AI red teaming practices.

Files changed

src/data/roadmaps/ai-red-teaming/ai-red-teaming.md

2025-04-26T11:56:24 See commit

This commit addresses the issue of broken URLs in the document titled "Why Red Team AI Systems?" located in the AI red teaming roadmap. The changes include the removal of several resource links that were likely outdated or non-functional, resulting in a net deletion of eight lines and one addition. The core content of the document remains intact, emphasizing the unique risks associated with AI systems and the necessity of AI Red Teaming to identify vulnerabilities that traditional testing methods may overlook.

By streamlining the content, the commit aims to enhance clarity and focus on the importance of AI Red Teaming without the distraction of broken links. The remaining text effectively highlights the critical need for adversary-focused insights to ensure the safety and reliability of AI systems prior to their deployment.

Files changed

src/data/roadmaps/ai-red-teaming/content/why-red-team-ai-systems@fNTb9y3zs1HPYclAmu_Wv.md

2025-04-29T21:36:39 See commit

This commit introduces a comprehensive roadmap for AI agents, encompassing a wide array of topics essential for understanding and developing AI agents. The commit includes the addition of a JSON file and a markdown file that outline the roadmap, along with numerous content files that cover specific concepts, methodologies, and technologies related to AI agents. Each content file addresses different aspects of AI agent development, such as tool invocation, memory management, API requests, and safety measures, among others.

The new roadmap serves as a structured guide for developers and researchers interested in AI agents, providing insights into various components that contribute to their functionality and effectiveness. By organizing these topics into a cohesive framework, the commit aims to facilitate learning and exploration in the rapidly evolving field of AI, ensuring that users can easily access relevant information and resources to enhance their understanding and implementation of AI agents.

Files changed

src/data/roadmaps/ai-agents/ai-agents.json
src/data/roadmaps/ai-agents/ai-agents.md
src/data/roadmaps/ai-agents/content/acting--tool-invocation@sHYd4KsKlmw5Im3nQ19W8.md
src/data/roadmaps/ai-agents/content/agent-loop@Eih4eybuYB3C2So8K0AT3.md
src/data/roadmaps/ai-agents/content/anthropic-tool-use@1EZFbDHA5J5_5BPMLMxXb.md
src/data/roadmaps/ai-agents/content/api-requests@52qxjZILV-X1isup6dazC.md
src/data/roadmaps/ai-agents/content/autogen@7YtnQ9-KIvGPSpDzEDexl.md
src/data/roadmaps/ai-agents/content/basic-backend-development@VPI89s-m885r2YrXjYxdd.md
src/data/roadmaps/ai-agents/content/be-specific-in-what-you-want@qFKFM2qNPEN7EoD0V-1SM.md
src/data/roadmaps/ai-agents/content/bias--toxicity-guardrails@EyLo2j8IQsIK91SKaXkmK.md
src/data/roadmaps/ai-agents/content/chain-of-thought-cot@qwdh5pkBbrF8LKPxbZp4F.md
src/data/roadmaps/ai-agents/content/closed-weight-models@tJYmEDDwK0LtEux-kwp9B.md
src/data/roadmaps/ai-agents/content/code-execution--repl@mS0EVCkWuPN_GkVPng4A2.md
src/data/roadmaps/ai-agents/content/code-generation@PK8w31GlvtmAuU92sHaqr.md
src/data/roadmaps/ai-agents/content/context-windows@dyn1LSioema-Bf9lLTgUZ.md
src/data/roadmaps/ai-agents/content/creating-mcp-servers@1NXIN-Hbjl5rPy_mqxQYW.md
src/data/roadmaps/ai-agents/content/crewai@uFPJqgU4qGvZyxTv-osZA.md
src/data/roadmaps/ai-agents/content/dag-agents@Ep8RoZSy_Iq_zWXlGQLZo.md
src/data/roadmaps/ai-agents/content/data-analysis@wKYEaPWNsR30TIpHaxSsq.md
src/data/roadmaps/ai-agents/content/data-privacy--pii-redaction@rdlYBJNNyZUshzsJawME4.md
src/data/roadmaps/ai-agents/content/database-queries@sV1BnA2-qBnXoKpUn-8Ub.md
src/data/roadmaps/ai-agents/content/deepeval@0924QUH1wV7Mp-Xu0FAhF.md
src/data/roadmaps/ai-agents/content/email--slack--sms@qaNr5I-NQPnfrRH7ynGTl.md
src/data/roadmaps/ai-agents/content/embeddings-and-vector-search@UIm54UmICKgep6s8Itcyv.md
src/data/roadmaps/ai-agents/content/episodic-vs-semantic-memory@EfCCNqLMJpWKKtamUa5gK.md
src/data/roadmaps/ai-agents/content/file-system-access@BoJqZvdGam4cd6G6yK2IV.md
src/data/roadmaps/ai-agents/content/fine-tuning-vs-prompt-engineering@5OW_6o286mj470ElFyJ_5.md
src/data/roadmaps/ai-agents/content/forgetting--aging-strategies@m-97m7SI0XpBnhEE8-_1S.md
src/data/roadmaps/ai-agents/content/frequency-penalty@z_N-Y0zGkv8_qHPuVtimL.md
src/data/roadmaps/ai-agents/content/gemini-function-calling@_iIsBJTVS6OBf_dsdmbVO.md
src/data/roadmaps/ai-agents/content/git-and-terminal-usage@McREk2zHOlIrqbGSKbX-J.md
src/data/roadmaps/ai-agents/content/haystack@XS-FsvtrXGZ8DPrwOsnlI.md
src/data/roadmaps/ai-agents/content/helicone@MLxP5N0Vrmwh-kyvNeGXn.md
src/data/roadmaps/ai-agents/content/human-in-the-loop-evaluation@rHxdxN97ZcU7MPl8L1jzN.md
src/data/roadmaps/ai-agents/content/integration-testing-for-flows@P9-SiIda3TSjHsfkI5OUV.md
src/data/roadmaps/ai-agents/content/iterate-and-test-your-prompts@noTuUFnHSBzn7GKG9UZEi.md
src/data/roadmaps/ai-agents/content/langchain@Ka6VpCEnqABvwiF9vba7t.md
src/data/roadmaps/ai-agents/content/langfuse@UoIheaJlShiceafrWALEH.md
src/data/roadmaps/ai-agents/content/langsmith@SS8mGqf9wfrNqenIWvN8Z.md
src/data/roadmaps/ai-agents/content/llamaindex@iEHF-Jm3ck-Iu85EbCoDi.md
src/data/roadmaps/ai-agents/content/llm-native-function-calling@aafZxtjxiwzJH1lwHBODi.md
src/data/roadmaps/ai-agents/content/local-desktop@iBtJp24F_kJE3YlBsW60s.md
src/data/roadmaps/ai-agents/content/long-term-memory@Ue633fz6Xu2wa2-KOAtdP.md
src/data/roadmaps/ai-agents/content/manual-from-scratch@US6T5dXM8IY9V2qZnTOFW.md
src/data/roadmaps/ai-agents/content/max-length@Bn_BkthrVX_vOuwQzvPZa.md
src/data/roadmaps/ai-agents/content/mcp-client@CGVstUxVXLJcYZrwk3iNQ.md
src/data/roadmaps/ai-agents/content/mcp-hosts@9FryAIrWRHh8YlzKX3et5.md
src/data/roadmaps/ai-agents/content/mcp-servers@yv_-87FVM7WKn5iv6LW9q.md
src/data/roadmaps/ai-agents/content/metrics-to-track@v8qLnyFRnEumodBYxQSXQ.md
src/data/roadmaps/ai-agents/content/model-context-protocol-mcp@1B0IqRNYdtbHDi1jHSXuI.md
src/data/roadmaps/ai-agents/content/npc--game-ai@ok8vN7VtCgyef5x6aoQaL.md
src/data/roadmaps/ai-agents/content/observation--reflection@ZJTrun3jK3zBGOTm1jdMI.md
src/data/roadmaps/ai-agents/content/open-weight-models@DSJAhQhc1dQmBHQ8ZkTau.md
src/data/roadmaps/ai-agents/content/openai-assistant-api@37GBFVZ2J2d5r8bd1ViHq.md
src/data/roadmaps/ai-agents/content/openai-functions-calling@AQtxTTxmBpfl8BMgJbGzc.md
src/data/roadmaps/ai-agents/content/openllmetry@7UqPXUzqKYXklnB3x-tsv.md
src/data/roadmaps/ai-agents/content/perception--user-input@LU76AhCYDjxdBhpMQ4eMU.md
src/data/roadmaps/ai-agents/content/personal-assistant@PPdAutqJF5G60Eg9lYBND.md
src/data/roadmaps/ai-agents/content/planner-executor@6YLCMWzystao6byCYCTPO.md
src/data/roadmaps/ai-agents/content/presence-penalty@Vd8ycw8pW-ZKvg5WYFtoh.md
src/data/roadmaps/ai-agents/content/pricing-of-common-models@B8dzg61TGaknuruBgkEJd.md
src/data/roadmaps/ai-agents/content/prompt-injection--jailbreaks@SU2RuicMUo8tiAsQtDI1k.md
src/data/roadmaps/ai-agents/content/provide-additional-context@6I42CoeWX-kkFXTKAY7rw.md
src/data/roadmaps/ai-agents/content/rag-agent@cW8O4vLLKEG-Q0dE8E5Zp.md
src/data/roadmaps/ai-agents/content/rag-and-vector-databases@wkS4yOJ3JdZQE_yBID8K7.md
src/data/roadmaps/ai-agents/content/ragas@YzEDtGEaMaMWVt0W03HRt.md
src/data/roadmaps/ai-agents/content/react-reason--act@53xDks6JQ33fHMa3XcuCd.md
src/data/roadmaps/ai-agents/content/reason-and-plan@ycPRgRYR4lEBQr_xxHKnM.md
src/data/roadmaps/ai-agents/content/reasoning-vs-standard-models@N3yZfUxphxjiupqGpyaS9.md
src/data/roadmaps/ai-agents/content/remote--cloud@dHNMX3_t1KSDdAWqgdJXv.md
src/data/roadmaps/ai-agents/content/rest-api-knowledge@QtTwecLdvQa8pgELJ6i80.md
src/data/roadmaps/ai-agents/content/safety--red-team-testing@63nsfJFO1BwjLX_ZVaPFC.md
src/data/roadmaps/ai-agents/content/short-term--memory@M3U6RfIqaiut2nuOibY8W.md
src/data/roadmaps/ai-agents/content/smol-depot@eWxQiBrxIUG2JNcrdfIHS.md
src/data/roadmaps/ai-agents/content/specify-length-format-etc@wwHHlEoPAx0TLxbtY6nMA.md
src/data/roadmaps/ai-agents/content/stopping-criteria@K0G-Lw069jXUJwZqHtybd.md
src/data/roadmaps/ai-agents/content/streamed-vs-unstreamed-responses@i2NE6haX9-7mdoV5LQ3Ah.md
src/data/roadmaps/ai-agents/content/structured-logging--tracing@zs6LM8WEnb0ERWpiaQCgc.md
src/data/roadmaps/ai-agents/content/summarization--compression@jTDC19BTWCqxqMizrIJHr.md
src/data/roadmaps/ai-agents/content/temperature@L1zL1GzqjSAjF06pIIXhy.md
src/data/roadmaps/ai-agents/content/token-based-pricing@1fiWPBV99E2YncqdCgUw2.md
src/data/roadmaps/ai-agents/content/tokenization@GAjuWyJl9CI1nqXBp6XCf.md
src/data/roadmaps/ai-agents/content/tool-definition@qakbxB8xe7Y8gejC5cZnK.md
src/data/roadmaps/ai-agents/content/tool-sandboxing--permissioning@UVzLGXG6K7HQVHmw8ZAv2.md
src/data/roadmaps/ai-agents/content/top-p@icbp1NjurQfdM0dHnz6v2.md
src/data/roadmaps/ai-agents/content/transformer-models-and-llms@ZF5_5Y5zqa75Ov22JACX6.md
src/data/roadmaps/ai-agents/content/tree-of-thought@Nmy1PoB32DcWZnPM8l8jT.md
src/data/roadmaps/ai-agents/content/understand-the-basics-of-rag@qwVQOwBTLA2yUgRISzC8k.md
src/data/roadmaps/ai-agents/content/unit-testing-for-individual-tools@qo_O4YAe4-MTP_ZJoXJHR.md
src/data/roadmaps/ai-agents/content/use-examples-in-your-prompt@yulzE4ZNLhXOgHhG7BtZQ.md
src/data/roadmaps/ai-agents/content/use-relevant-technical-terms@sUwdtOX550tSdceaeFPmF.md
src/data/roadmaps/ai-agents/content/user-profile-storage@QJqXHV8VHPTnfYfmKPzW7.md
src/data/roadmaps/ai-agents/content/web-scraping--crawling@5oLc-235bvKhApxzYFkEc.md
src/data/roadmaps/ai-agents/content/web-search@kBtqT8AduLoYDWopj-V9_.md
src/data/roadmaps/ai-agents/content/what-are-ai-agents@aFZAm44nP5NefX_9TpT0A.md
src/data/roadmaps/ai-agents/content/what-are-tools@2zsOUWJQ8e7wnoHmq1icG.md
src/data/roadmaps/ai-agents/content/what-is-agent-memory@TBH_DZTAfR8Daoh-njNFC.md
src/data/roadmaps/ai-agents/content/what-is-prompt-engineering@Y8EqzFx3qxtrSh7bWbbV8.md

2025-04-29T22:00:38 See commit

This commit introduces updates to the AI Agents roadmap, specifically in the file located at src/data/roadmaps/ai-agents/ai-agents.md. The primary changes include a revision of the brief and detailed descriptions to emphasize the design, building, and shipping of AI agents in 2025, rather than simply learning about them. The roadmap is marked as new and visible, and it retains its structured format with dimensions and schema information.

Additionally, the commit modifies the roadmap's metadata, ensuring that the descriptions align with the new focus on practical application and resource accessibility. The roadmap is intended to serve as an interactive guide, providing users with a comprehensive learning experience that includes resources and short descriptions related to each item. The updates reflect a shift towards a more hands-on approach in learning about AI agents.

Files changed

src/data/roadmaps/ai-agents/ai-agents.md

2025-04-29T22:16:46 See commit

This commit focuses on enhancing the AI Agents roadmap by updating multiple files to include new content and information. Key modifications were made to the readme.md, roadmap-tree-content.js, and a comprehensive set of JSON and markdown files that detail various aspects of AI agents, such as their functionalities, tools, and methodologies. Each modified file corresponds to specific topics within the AI agents domain, reflecting an effort to provide clearer guidance and resources for developers and users interested in AI technologies.

The updates span a wide range of subjects, including tool invocation, agent loops, data privacy, and memory management, among others. By refining these documents, the commit aims to improve the overall understanding and usability of AI agents, ensuring that users have access to the latest information and best practices in the field. This structured approach not only enhances the roadmap's clarity but also supports ongoing development efforts in AI technologies.

Files changed

readme.md
scripts/roadmap-tree-content.js
src/data/roadmaps/ai-agents/ai-agents.json
src/data/roadmaps/ai-agents/content/acting--tool-invocation@sHYd4KsKlmw5Im3nQ19W8.md
src/data/roadmaps/ai-agents/content/agent-loop@Eih4eybuYB3C2So8K0AT3.md
src/data/roadmaps/ai-agents/content/anthropic-tool-use@1EZFbDHA5J5_5BPMLMxXb.md
src/data/roadmaps/ai-agents/content/api-requests@52qxjZILV-X1isup6dazC.md
src/data/roadmaps/ai-agents/content/autogen@7YtnQ9-KIvGPSpDzEDexl.md
src/data/roadmaps/ai-agents/content/be-specific-in-what-you-want@qFKFM2qNPEN7EoD0V-1SM.md
src/data/roadmaps/ai-agents/content/bias--toxicity-guardrails@EyLo2j8IQsIK91SKaXkmK.md
src/data/roadmaps/ai-agents/content/chain-of-thought-cot@qwdh5pkBbrF8LKPxbZp4F.md
src/data/roadmaps/ai-agents/content/closed-weight-models@tJYmEDDwK0LtEux-kwp9B.md
src/data/roadmaps/ai-agents/content/code-execution--repl@mS0EVCkWuPN_GkVPng4A2.md
src/data/roadmaps/ai-agents/content/code-generation@PK8w31GlvtmAuU92sHaqr.md
src/data/roadmaps/ai-agents/content/context-windows@dyn1LSioema-Bf9lLTgUZ.md
src/data/roadmaps/ai-agents/content/creating-mcp-servers@1NXIN-Hbjl5rPy_mqxQYW.md
src/data/roadmaps/ai-agents/content/crewai@uFPJqgU4qGvZyxTv-osZA.md
src/data/roadmaps/ai-agents/content/dag-agents@Ep8RoZSy_Iq_zWXlGQLZo.md
src/data/roadmaps/ai-agents/content/data-analysis@wKYEaPWNsR30TIpHaxSsq.md
src/data/roadmaps/ai-agents/content/data-privacy--pii-redaction@rdlYBJNNyZUshzsJawME4.md
src/data/roadmaps/ai-agents/content/database-queries@sV1BnA2-qBnXoKpUn-8Ub.md
src/data/roadmaps/ai-agents/content/deepeval@0924QUH1wV7Mp-Xu0FAhF.md
src/data/roadmaps/ai-agents/content/email--slack--sms@qaNr5I-NQPnfrRH7ynGTl.md
src/data/roadmaps/ai-agents/content/embeddings-and-vector-search@UIm54UmICKgep6s8Itcyv.md
src/data/roadmaps/ai-agents/content/episodic-vs-semantic-memory@EfCCNqLMJpWKKtamUa5gK.md
src/data/roadmaps/ai-agents/content/file-system-access@BoJqZvdGam4cd6G6yK2IV.md
src/data/roadmaps/ai-agents/content/fine-tuning-vs-prompt-engineering@5OW_6o286mj470ElFyJ_5.md
src/data/roadmaps/ai-agents/content/forgetting--aging-strategies@m-97m7SI0XpBnhEE8-_1S.md
src/data/roadmaps/ai-agents/content/frequency-penalty@z_N-Y0zGkv8_qHPuVtimL.md
src/data/roadmaps/ai-agents/content/gemini-function-calling@_iIsBJTVS6OBf_dsdmbVO.md
src/data/roadmaps/ai-agents/content/haystack@XS-FsvtrXGZ8DPrwOsnlI.md
src/data/roadmaps/ai-agents/content/helicone@MLxP5N0Vrmwh-kyvNeGXn.md
src/data/roadmaps/ai-agents/content/human-in-the-loop-evaluation@rHxdxN97ZcU7MPl8L1jzN.md
src/data/roadmaps/ai-agents/content/integration-testing-for-flows@P9-SiIda3TSjHsfkI5OUV.md
src/data/roadmaps/ai-agents/content/iterate-and-test-your-prompts@noTuUFnHSBzn7GKG9UZEi.md
src/data/roadmaps/ai-agents/content/langchain@Ka6VpCEnqABvwiF9vba7t.md
src/data/roadmaps/ai-agents/content/langfuse@UoIheaJlShiceafrWALEH.md
src/data/roadmaps/ai-agents/content/langsmith@SS8mGqf9wfrNqenIWvN8Z.md
src/data/roadmaps/ai-agents/content/langsmith@xp7TCTRE9HP60_rGzTUF6.md
src/data/roadmaps/ai-agents/content/llamaindex@iEHF-Jm3ck-Iu85EbCoDi.md
src/data/roadmaps/ai-agents/content/llm-native-function-calling@aafZxtjxiwzJH1lwHBODi.md
src/data/roadmaps/ai-agents/content/local-desktop@iBtJp24F_kJE3YlBsW60s.md
src/data/roadmaps/ai-agents/content/long-term-memory@Ue633fz6Xu2wa2-KOAtdP.md
src/data/roadmaps/ai-agents/content/manual-from-scratch@US6T5dXM8IY9V2qZnTOFW.md
src/data/roadmaps/ai-agents/content/max-length@Bn_BkthrVX_vOuwQzvPZa.md
src/data/roadmaps/ai-agents/content/mcp-client@CGVstUxVXLJcYZrwk3iNQ.md
src/data/roadmaps/ai-agents/content/mcp-hosts@9FryAIrWRHh8YlzKX3et5.md
src/data/roadmaps/ai-agents/content/mcp-servers@yv_-87FVM7WKn5iv6LW9q.md
src/data/roadmaps/ai-agents/content/metrics-to-track@v8qLnyFRnEumodBYxQSXQ.md
src/data/roadmaps/ai-agents/content/model-context-protocol-mcp@1B0IqRNYdtbHDi1jHSXuI.md
src/data/roadmaps/ai-agents/content/npc--game-ai@ok8vN7VtCgyef5x6aoQaL.md
src/data/roadmaps/ai-agents/content/observation--reflection@ZJTrun3jK3zBGOTm1jdMI.md
src/data/roadmaps/ai-agents/content/open-weight-models@DSJAhQhc1dQmBHQ8ZkTau.md
src/data/roadmaps/ai-agents/content/openai-assistant-api@37GBFVZ2J2d5r8bd1ViHq.md
src/data/roadmaps/ai-agents/content/openai-functions-calling@AQtxTTxmBpfl8BMgJbGzc.md
src/data/roadmaps/ai-agents/content/openllmetry@7UqPXUzqKYXklnB3x-tsv.md
src/data/roadmaps/ai-agents/content/perception--user-input@LU76AhCYDjxdBhpMQ4eMU.md
src/data/roadmaps/ai-agents/content/personal-assistant@PPdAutqJF5G60Eg9lYBND.md
src/data/roadmaps/ai-agents/content/planner-executor@6YLCMWzystao6byCYCTPO.md
src/data/roadmaps/ai-agents/content/presence-penalty@Vd8ycw8pW-ZKvg5WYFtoh.md
src/data/roadmaps/ai-agents/content/pricing-of-common-models@B8dzg61TGaknuruBgkEJd.md
src/data/roadmaps/ai-agents/content/prompt-injection--jailbreaks@SU2RuicMUo8tiAsQtDI1k.md
src/data/roadmaps/ai-agents/content/provide-additional-context@6I42CoeWX-kkFXTKAY7rw.md
src/data/roadmaps/ai-agents/content/rag-agent@cW8O4vLLKEG-Q0dE8E5Zp.md
src/data/roadmaps/ai-agents/content/rag-and-vector-databases@wkS4yOJ3JdZQE_yBID8K7.md
src/data/roadmaps/ai-agents/content/ragas@YzEDtGEaMaMWVt0W03HRt.md
src/data/roadmaps/ai-agents/content/react-reason--act@53xDks6JQ33fHMa3XcuCd.md
src/data/roadmaps/ai-agents/content/reason-and-plan@ycPRgRYR4lEBQr_xxHKnM.md
src/data/roadmaps/ai-agents/content/reasoning-vs-standard-models@N3yZfUxphxjiupqGpyaS9.md
src/data/roadmaps/ai-agents/content/remote--cloud@dHNMX3_t1KSDdAWqgdJXv.md
src/data/roadmaps/ai-agents/content/safety--red-team-testing@63nsfJFO1BwjLX_ZVaPFC.md
src/data/roadmaps/ai-agents/content/short-term--memory@M3U6RfIqaiut2nuOibY8W.md
src/data/roadmaps/ai-agents/content/smol-depot@eWxQiBrxIUG2JNcrdfIHS.md
src/data/roadmaps/ai-agents/content/specify-length-format-etc@wwHHlEoPAx0TLxbtY6nMA.md
src/data/roadmaps/ai-agents/content/stopping-criteria@K0G-Lw069jXUJwZqHtybd.md
src/data/roadmaps/ai-agents/content/streamed-vs-unstreamed-responses@i2NE6haX9-7mdoV5LQ3Ah.md
src/data/roadmaps/ai-agents/content/structured-logging--tracing@zs6LM8WEnb0ERWpiaQCgc.md
src/data/roadmaps/ai-agents/content/summarization--compression@jTDC19BTWCqxqMizrIJHr.md
src/data/roadmaps/ai-agents/content/temperature@L1zL1GzqjSAjF06pIIXhy.md
src/data/roadmaps/ai-agents/content/token-based-pricing@1fiWPBV99E2YncqdCgUw2.md
src/data/roadmaps/ai-agents/content/tokenization@GAjuWyJl9CI1nqXBp6XCf.md
src/data/roadmaps/ai-agents/content/tool-definition@qakbxB8xe7Y8gejC5cZnK.md
src/data/roadmaps/ai-agents/content/tool-sandboxing--permissioning@UVzLGXG6K7HQVHmw8ZAv2.md
src/data/roadmaps/ai-agents/content/top-p@icbp1NjurQfdM0dHnz6v2.md
src/data/roadmaps/ai-agents/content/transformer-models-and-llms@ZF5_5Y5zqa75Ov22JACX6.md
src/data/roadmaps/ai-agents/content/tree-of-thought@Nmy1PoB32DcWZnPM8l8jT.md
src/data/roadmaps/ai-agents/content/tree-of-thought@hj1adjkG9nalXKZ-Youn0.md
src/data/roadmaps/ai-agents/content/understand-the-basics-of-rag@qwVQOwBTLA2yUgRISzC8k.md
src/data/roadmaps/ai-agents/content/unit-testing-for-individual-tools@qo_O4YAe4-MTP_ZJoXJHR.md
src/data/roadmaps/ai-agents/content/use-examples-in-your-prompt@yulzE4ZNLhXOgHhG7BtZQ.md
src/data/roadmaps/ai-agents/content/use-relevant-technical-terms@sUwdtOX550tSdceaeFPmF.md
src/data/roadmaps/ai-agents/content/user-profile-storage@QJqXHV8VHPTnfYfmKPzW7.md
src/data/roadmaps/ai-agents/content/web-scraping--crawling@5oLc-235bvKhApxzYFkEc.md
src/data/roadmaps/ai-agents/content/web-search@kBtqT8AduLoYDWopj-V9_.md
src/data/roadmaps/ai-agents/content/what-are-ai-agents@aFZAm44nP5NefX_9TpT0A.md
src/data/roadmaps/ai-agents/content/what-are-tools@2zsOUWJQ8e7wnoHmq1icG.md
src/data/roadmaps/ai-agents/content/what-is-agent-memory@TBH_DZTAfR8Daoh-njNFC.md
src/data/roadmaps/ai-agents/content/what-is-prompt-engineering@Y8EqzFx3qxtrSh7bWbbV8.md

2025-04-29T22:18:44 See commit

This commit introduces a new roadmap for "AI Agents" on both the get-started and roadmaps pages of the project. Specifically, it modifies the RoadmapsPage.tsx file to include a new entry under the roles section, linking to the AI Agents page. Additionally, the get-started.astro file has been updated to add a new RoadmapCard that describes the AI Agents roadmap, highlighting a focus on designing, building, and shipping AI agents by 2025.

The changes enhance the project's navigation and provide users with clear guidance on pursuing knowledge in AI Agents, thereby expanding the educational resources available to those interested in this emerging field. The addition of this roadmap reflects an ongoing commitment to keep the content relevant and informative for users looking to engage with AI technologies.

Files changed

src/components/Roadmaps/RoadmapsPage.tsx
src/pages/get-started.astro

2025-04-29T22:20:08 See commit

This commit addresses an adjustment to the height specification in the AI Agents roadmap file located at src/data/roadmaps/ai-agents/ai-agents.md. The height has been modified from 3300 to 3230, reflecting a minor correction in the roadmap's dimensions.

Alongside this height adjustment, the commit retains the overall structure of the roadmap, which serves as an interactive guide for designing, building, and deploying AI agents in 2025. The roadmap continues to include resources and descriptions for each item, ensuring users have access to comprehensive learning materials in one convenient location.

Files changed

src/data/roadmaps/ai-agents/ai-agents.md

2025-04-30T11:27:24 See commit

This commit involves updates to two markdown files related to AI agents, specifically focusing on the concepts of episodic vs. semantic memory and short-term memory. In the first file, the description of episodic memory has been modified to emphasize its relevance to the current conversation context and its transient nature, contrasting it with semantic memory, which is characterized by its persistent and broader knowledge scope. The changes clarify that episodic memory may be lost after a conversation ends, while semantic memory retains timeless knowledge.

In the second file, the definition of short-term memory has been expanded significantly. The updated content explains that short-term memory consists of facts incorporated into prompts for large language models (LLMs), providing a structured example of how user profiles, current topics, and goals can be included as part of the memory. This elaboration highlights the practical application of short-term memory in facilitating user interactions and goal achievement, while also outlining design considerations for maintaining essential details without overwhelming the system with unnecessary data.

Files changed

src/data/roadmaps/ai-agents/content/episodic-vs-semantic-memory@EfCCNqLMJpWKKtamUa5gK.md
src/data/roadmaps/ai-agents/content/short-term--memory@M3U6RfIqaiut2nuOibY8W.md

2025-04-30T15:31:44 See commit

The commit titled "chore: update roadmap content json (#8567)" involves updates to several JSON files within the project's roadmap content. The changes include the addition of a new file for AI agents, as well as modifications to existing files related to AI data scientists, AI red teaming, and cyber security.

Co-authored by Kamran Ahmed, this update reflects ongoing efforts to enhance and refine the project's roadmap, ensuring that the content remains current and relevant for users. The specific modifications made to the existing files likely involve updates to their content or structure, further aligning them with the project's goals and objectives.

Files changed

public/roadmap-content/ai-agents.json
public/roadmap-content/ai-data-scientist.json
public/roadmap-content/ai-red-teaming.json
public/roadmap-content/cyber-security.json

2025-04-30T15:32:12 See commit

The commit addresses a syntax error in the use-custom-roadmap.ts file by enhancing error handling within the useCustomRoadmap hook. It modifies the query function to wrap the URL construction and HTTP request in a try-catch block. If a SyntaxError occurs during the process, it throws a custom FetchError with a 404 status, indicating that the roadmap was not found. Additionally, it ensures that the roadmapUrl is constructed correctly based on the presence of a slug or id, and includes the secret parameter if applicable.

In the query-http.ts file, the FetchError interface is transformed into a class, allowing for more structured error handling. The constructor of the FetchError class initializes the error's status and message, and a static method is added to check if an error is an instance of FetchError. This change improves the overall error handling mechanism, ensuring that specific error types can be caught and managed appropriately throughout the application.

Files changed

src/hooks/use-custom-roadmap.ts
src/lib/query-http.ts