Skip to content
View Elon-Jiang's full-sized avatar
3 followers · 1 following
  • HangZhou

Achievements

Achievement: Pull Sharkx2

Achievements

Achievement: Pull Sharkx2

Block or report Elon-Jiang

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Elon-Jiang/README.md
Elon Jiang — Building agent harnesses, MCP tooling, and AI applications. Active open-source contributor. 38 PRs merged · 64 open

107 public PRs · 38 merged / 64 open / 5 closed

Counts only public PRs I authored in repositories owned by others. Stars belong to the contributed projects.

Adopted upstream

Contributions incorporated into the upstream main branch through maintainer sync, rather than GitHub PR merges. Listed separately from the merged PR count.

All 107 PRs and repositories
Repository Pull request Status
Tencent/AI-Infra-Guard #761 fix(runner): stop CIDR enumeration at final IPv4 address Open
Tencent/AI-Infra-Guard #760 fix(agent-scan,mcp-scan): honor finish tool name casing Open
agentscope-ai/OpenJudge #238 fix(text): compare numbers before float rounding Open
agentscope-ai/OpenJudge #237 fix(format): keep soft penalty zero at threshold Open
getzep/graphiti #2148 fix: bind positional driver for single FalkorDB group Open
e2b-dev/E2B #2137 fix(sdk): resolve directory symlinks in template copies Open
a2aproject/a2a-python #1464 fix(client): handle case-insensitive SSE media types Open
langchain-ai/langgraphjs #3368 fix(langgraph-checkpoint-postgres): honor empty membership filters Open
567-labs/instructor #3037 fix(partial): preserve typed tuples in streamed snapshots Open
e2b-dev/E2B #2135 fix(template): preserve millisecond readiness waits Open
deepset-ai/haystack #13690 fix(tools): keep dataclass default factories lazy Open
Tencent/AI-Infra-Guard #755 fix(scanners): handle async local tool execution errors Merged
Tencent/AI-Infra-Guard #754 fix(agent-scan): decode batch arrays from XML parameters Merged
agentscope-ai/OpenJudge #234 fix(grader): await results from callable evaluators Merged
agentscope-ai/OpenJudge #233 fix(runner): accept a single callable aggregator Merged
NVIDIA/Isaac-GR00T #810 fix(policy): close retired sockets during timeout recovery Closed
AtsushiSakai/PythonRobotics #1515 Preserve clean controls in histogram localization observations Open
Farama-Foundation/Gymnasium-Robotics #365 Reject impossible maze reset sampling instead of looping forever Open
AtsushiSakai/PythonRobotics #1514 Fix FastSLAM resampling state and landmark copies Open
petercorke/robotics-toolbox-python #937 fix(mobile): preserve fractional limits for integer controls Open
Physical-Intelligence/openpi #1087 Fix query-sensitive download caching and literal file retrieval Open
NVIDIA/Isaac-GR00T #809 Fix consecutive non-monotonic timestamps in action interpolation Closed
RoboVerseOrg/RoboVerse #886 fix(camera): construct valid batched and vertical views Open
RoboVerseOrg/RoboVerse #885 fix(replay): fail validation on non-finite states Open
Farama-Foundation/Gymnasium-Robotics #364 Fix FrankaKitchen reward recomputation after task completion Open
VoltAgent/voltagent #1524 fix(core): report in-memory file sizes in UTF-8 bytes Open
Tencent/AI-Infra-Guard #747 fix(agent-scan): preserve prompt escapes in provider templates Merged
agentscope-ai/OpenJudge #229 fix(models): honor per-call OpenAI streaming mode Merged
Tencent/AI-Infra-Guard #746 fix(agent-scan): preserve non-object HTTP error responses Merged
agentscope-ai/OpenJudge #228 fix(models): format Omni audio once Merged
Portkey-AI/gateway #1897 bug: preserve Cohere base64 embedding results Open
567-labs/instructor #2992 fix(templating): render an explicitly supplied empty context Open
Portkey-AI/gateway #1896 bug: preserve UTF-8 across JSONL stream chunks Open
assafelovic/gpt-researcher #2275 fix(markdown): keep fenced code out of report headings Open
HKUDS/RAG-Anything #469 fix(resilience): release interrupted synchronous recovery probes Merged
assistant-ui/assistant-ui #10397 fix(react-a2a): merge protocol header names case-insensitively Merged
VoltAgent/voltagent #1519 fix(rag): keep NeuralChunker output IDs distinct after expansion Open
VoltAgent/voltagent #1517 fix(sdk): preserve request header forms and overrides Open
assistant-ui/assistant-ui #10395 fix(assistant-stream): stop scanning SSE input at the done marker Merged
neuml/txtai #1531 Fix search worker limits when CPU count is unknown Open
vibrantlabsai/ragas #3098 fix: discover Git worktrees for experiment versioning Open
neuml/txtai #1529 Fix iterator outputs in batched workflow tasks Open
vibrantlabsai/ragas #3096 fix(dataset): keep seeded train/test splits reproducible Open
run-llama/llama_index #24065 fix(core): preserve chat memory input snapshots during restore Open
run-llama/llama_index #23976 fix(core): isolate SimpleKVStore dictionary snapshots Open
langchain-ai/deepagents #7560 fix(sdk): report decoded binary file sizes Closed
MervinPraison/PraisonAI #6199 fix(ts): interrupt tool retry backoff on cancellation Merged
MervinPraison/PraisonAI #6198 fix(ts): enforce tool payload limits in UTF-8 bytes Merged
MervinPraison/PraisonAI #6197 fix(ts): invalidate tool installation cache on registration changes Merged
MervinPraison/PraisonAI #6196 fix(ts): re-enter downstream tool middleware on retry Merged
MervinPraison/PraisonAI #6195 fix(ts): release hook deadline timers after settlement Merged
MervinPraison/PraisonAI #6194 fix(ts): keep hook inspection from freezing registration Merged
MervinPraison/PraisonAI #6193 fix(ts): preserve concurrent FileCache list appends Merged
MervinPraison/PraisonAI #6192 fix(ts): initialize and route MCP stdio requests correctly Open
MervinPraison/PraisonAI #6191 fix(ts): remove stale graph neighbors when an edge is rewired Merged
MervinPraison/PraisonAI #6190 fix(ts): ensure fixed-size chunking always advances Open
MervinPraison/PraisonAI #6188 fix(ts): preserve sentence chunk content and source offsets Merged
MervinPraison/PraisonAI #6186 fix(ts): preserve session changes during file saves Merged
genkit-ai/genkit #7000 fix(py): isolate generated output format metadata Open
genkit-ai/genkit #6999 fix(py): preserve message metadata when injecting output instructions Open
algorithmicsuperintelligence/openevolve #578 fix(llm): stop Codex process before closing read pipes Merged
algorithmicsuperintelligence/openevolve #577 fix(llm): use UTF-8 for model CLI text pipes Merged
algorithmicsuperintelligence/openevolve #576 fix(llm): reject failed Claude CLI responses before returning stdout Merged
algorithmicsuperintelligence/openevolve #575 fix(trace): keep JSON export metadata reusable Merged
MervinPraison/PraisonAI #6184 fix(ts): settle MCP requests when WebSocket connections close Merged
MervinPraison/PraisonAI #6183 fix(ts): evict empty keys from bounded MemoryCache Merged
MervinPraison/PraisonAI #6180 fix(ts): honor the configured FileCache TTL Merged
genkit-ai/genkit #6998 fix(py): handle recursive schemas in array and JSONL formats Open
MervinPraison/PraisonAI #6178 fix(ts): resolve MCP WebSocket responses with zero request IDs Merged
assafelovic/gpt-researcher #2252 fix(document): detect online document type from URL path Merged
genkit-ai/genkit #6996 fix(py): ignore SSE heartbeat comments in agent streams Open
genkit-ai/genkit #6995 fix(py): compare session snapshot creation instants chronologically Open
simular-ai/Agent-S #253 fix(s4): quote paths when archiving interface variants Open
NevaMind-AI/memU #775 fix(embedding): skip empty SDK embedding requests Open
genkit-ai/genkit #6992 fix(py): preserve application import errors in run_main Open
genkit-ai/genkit #6991 fix(py): preserve nullable chunks in waiting stream readers Open
assafelovic/gpt-researcher #2250 fix(vector-store): skip empty chunk batches Merged
2FastLabs/agent-squad #726 fix(python): preserve memory and SQL message timestamps Open
2FastLabs/agent-squad #724 fix(storage): persist raw messages before updating summaries Open
agentscope-ai/OpenJudge #204 fix(runner): reject nonpositive concurrency Merged
VoltAgent/voltagent #1492 fix(core): snapshot in-memory persistence payloads Open
VoltAgent/voltagent #1491 fix(core): scope composite filesystem searches to requested paths Open
langchain-ai/langchainjs #12077 fix(core): preserve prototype property names in RunnableMap Open
algorithmicsuperintelligence/openevolve #568 fix(llm): avoid assuming custom clients store a model field Merged
MemoriLabs/Memori #650 fix(llm): preserve synchronous streaming during client registration Open
HKUDS/RAG-Anything #454 fix(resilience): await all asynchronous retry callback results Merged
HKUDS/LightRAG #4288 fix(utils): preserve cosine direction across finite vector scales Open
microsoft/agent-lightning #667 fix(controller): await non-coroutine local worker results Closed
browser-use/browser-use #6260 fix(tokens): preserve concurrent pricing cache snapshots Open
HKUDS/LightRAG #4287 fix(vlm): accept case-insensitive image data URLs Open
HKUDS/DeepCode #288 fix(paper2code): preserve fenced code in paper sections Open
browserbase/stagehand #3360 fix(sdk-ts): observe rejected operations after pre-abort Open
MervinPraison/PraisonAI #6149 fix(ts): preserve arbitrary session state keys Merged
MervinPraison/PraisonAI #6147 fix(ts): reject negative memory capacities Closed
e2b-dev/E2B #2072 fix(python-sdk): notify on_exit after immediate watch stop Open
agentscope-ai/OpenJudge #203 fix(analyzer): restore consistency call support Merged
confident-ai/deepeval #3560 fix(classifiers): match whitespace-padded label names Open
browserbase/stagehand #3355 fix(sdk-ts): reject FIFO uploads without blocking Open
VoltAgent/voltagent #1490 fix(evals): propagate experiment callback failures Open
Tencent/AI-Infra-Guard #733 fix(mcp-scan): normalize mixed-case MCP SARIF rule IDs Merged
HKUDS/DeepCode #287 fix(memory): keep distinct topic pointers during consolidation Open
CopilotKit/CopilotKit #8114 fix(core): handle inherited property names in thread store registry Open
MervinPraison/PraisonAI #6142 fix(ts): recover FileMemory writes after append failures Merged
CopilotKit/CopilotKit #8111 fix(core): replace all case variants when overlaying headers Open
TencentCloud/Octop #2625 fix(dashboard): display custom ACP runner names and commands Open
TencentCloud/Octop #2624 fix(dashboard): keep HTML preview anchors inside the document Open
pydantic/pydantic-ai #11833 Allow Coder and Researcher to load from agent specs Draft

Refreshed daily · Update history

Popular repositories Loading

  1. robo-agent-bench robo-agent-bench Public

    Can an LLM agent actually tidy a table? Physics-scored household manipulation benchmark for LLM/VLM agents driving a simulated robot arm (MuJoCo).

    Python 1

  2. ultimate-react-course ultimate-react-course Public

    Forked from jonasschmedtmann/ultimate-react-course

    Starter files, final projects, and FAQ for my Ultimate React course

    JavaScript

  3. Python_Data_Analysis Python_Data_Analysis Public

    Python数据分析练习，包括数据读取、评估、清洗、分析、可视化

    Jupyter Notebook

  4. about about Public

  5. privacy privacy Public

  6. terms terms Public