models-resources-vector
Guides SuperAGI model providers, resource ingestion, knowledge flows, vector stores, embeddings, and storage configuration troubleshooting.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- models-resources-vector
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Guides SuperAGI model providers, resource ingestion, knowledge flows, vector stores, embeddings, and storage configuration troubleshooting.
- disable-model-invocation
- true
02 资源文件 5
- 参考文档references/model-provider-reference.md1.6 KB
- 参考文档references/resources-and-knowledge.md1.7 KB
- 参考文档references/troubleshooting.md1.7 KB
- 参考文档references/vector-store-reference.md2.0 KB
- 脚本scripts/check_provider_config.py4.0 KB
03 SKILL.md 正文
SuperAGI Models, Resources, and Vector Stores
Use this sub-skill when the task is about provider selection, model API keys, resource summarization, knowledge installation, vector DB configuration, or embedding/vector-store behavior.
Read First
- references/model-provider-reference.md for model source enums, provider factory behavior, and local LLM notes.
- references/resources-and-knowledge.md for resource storage, summarization, and knowledge flow.
- references/vector-store-reference.md for vector-store enums, factory coverage, and backend expectations.
- references/troubleshooting.md for provider, storage, credential, and vector failures.
- scripts/check_provider_config.py for a safe structural check of model/vector/resource config combinations.
Main Topics
- Model provider enums and factories: OpenAI, Google Palm, Replicate, Hugging Face, and Local LLM.
- Resource storage: FILE vs S3, plus resource summarization into vector stores.
- Knowledge records and vector DB indices.
- Vector store selection and factory paths for Redis, Pinecone, Weaviate, Qdrant, and related enums.
- Image LLM wrappers when a task names DALL·E or Stable Diffusion.
Safe Workflow
- Decide whether the user is asking about providers, resources, knowledge, or vector stores; use the specific reference rather than general troubleshooting.
- Check the config and credential requirements before assuming a backend is available.
- Treat provider-key validation and vector-store connection checks as live operations that may touch external services.
- Use the config checker for structural validation before running a live request.
Boundary Notes
- API route prefixes, controllers, and CRUD endpoints belong to
api-service. - Tool wrappers that call model/vector providers belong to
toolkits-integrations. - Deployment/runtime setup and config file creation belong to
deployment-configuration. - Agent execution logic that consumes the selected provider or vector store
belongs to
agents-workflows.
04 同仓库更多 Skills
passnet-triton-opt
Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.
passnet-solve
END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.
passnet-skill
PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.
passnet-pattern-fusion
Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.
passnet-orchestrate
Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.
passnet-feedback
Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).