bytedance/agentkit-samples is a collection of examples and tutorials for Volcengine AgentKit, an AI-agent development platform for building, deploying, and operating agent applications. Developers use the samples to learn agent creation, multi-agent collaboration, memory, retrieval, MCP integrations, media generation, customer service, and other workflows. The catalogue skills provide agent workflows based on these examples.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add bytedance/agentkit-samples --skill byted-milvusgit clone --depth 1 https://github.com/bytedance/agentkit-samplesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bytedance/agentkit-samples/byted-milvus)<a href="https://agentmods.dev/skills/bytedance/agentkit-samples/byted-milvus"><img src="https://agentmods.dev/badge/skills/bytedance/agentkit-samples/byted-milvus.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.01050 |
| Opus 5 | $0.00030 | $0.00525 |
| Sonnet 5 | $0.00012 | $0.00210 |
| Haiku 4.5 | $0.00006 | $0.00105 |
Grade A, and why
byted-milvus scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Volcano Engine Milvus
Manage Milvus instances on Volcano Engine — cluster lifecycle and vector data operations.
Quick Start
Use the bundled CLIs (always run via the skill venv):
{baseDir}/venv/bin/python {baseDir}/scripts/control.py <command>
{baseDir}/venv/bin/python {baseDir}/scripts/data.py <command>
If {baseDir}/venv does not exist:
python3 -m venv {baseDir}/venv
{baseDir}/venv/bin/pip install -r {baseDir}/requirements.txt
See CONTROL_PLANE.md and DATA_PLANE.md for workflows and examples.
Low-level control-plane fallback (use only when goal-based commands do not cover the task):
{baseDir}/venv/bin/python {baseDir}/scripts/control_tools.py <command>
Available operations
Control Plane (cluster management): Use goal-based workflows to provision, inspect, scale, delete, and expose Milvus instances.
→ See CONTROL_PLANE.md for goal-based commands and workflows.
→ See CONTROL_TOOLS.md for low-level control_tools.py fallback commands (use only when CONTROL_PLANE.md does not cover the task).
Data Plane (collections & data): Create/drop collections, insert/upsert/delete data, vector search, scalar query, and get-by-ID. → See DATA_PLANE.md for commands and use cases.
Out of scope
- Deploying or operating Milvus outside Volcano Engine (self-hosted, other clouds).
- Deep Milvus performance tuning or schema design beyond basic collection creation and queries.
- Application-level embedding strategy decisions (chunking, RAG design) unless needed to run the provided data plane commands.
Rules
Common
- Execution environment: Always use
{baseDir}/venv/bin/pythonto run scripts. - Authentication:
VOLCENGINE_ACCESS_KEYandVOLCENGINE_SECRET_KEYare required for all control-plane operations. Data-plane commands also require a reachable Milvus--endpointplus any needed Milvus auth flags. See DATA_PLANE.md. - Script usage: Prioritize
scripts/control.pyandscripts/data.py. Usescripts/control_tools.pyonly as a last resort when goal-based commands do not cover the task. Do not write ad-hoc Python scripts or use the SDK directly unless existing CLIs cannot satisfy a specific requirement. - Language (strict): Always reply in the user's language. Use a deterministic heuristic:
- If the user's message contains any Chinese characters, reply in Chinese.
- Otherwise, reply in the user's language as inferred from their message.
- Keep commands/flags/code in English; only the explanation and prompts should be localized.
- If the user mixes languages and preference is unclear, ask which language they prefer.
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- CONTROL_PLANE.md 9.3 KB
- CONTROL_TOOLS.md 7.2 KB
- DATA_PLANE.md 15 KB
- LICENSE 9.9 KB
- README.md 2.9 KB
- requirements.txt 85 B
- scripts/api.py 2.7 KB runs code
- scripts/ark_shim.py 2.0 KB runs code
- scripts/control_tools.py 30 KB runs code
- scripts/control.py 49 KB runs code
- scripts/data.py 37 KB runs code
- scripts/sdk_shim.py 2.4 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 72 lines · 0 tokens per session scan A c2d76cd8ce69
byted-milvus is a skill published in the GitHub repository bytedance/agentkit-samples (449 stars, last pushed 3d ago), licensed Apache-2.0. It adds 60 tokens to every session and 1,050 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
pinecone
Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense + sparse), metadata filtering, and namespaces. Low latency (<100ms p95). Use for production RAG, recommendation systems, or semantic search at scale. Best for serverless, managed infrastructure.
vector-db
Vector database expert for embeddings, similarity search, RAG patterns, and indexing strategies.
pinecone
Managed vector DB for production RAG and search.
bigquery-ai-ml
Skill for BigQuery AI and Machine Learning queries using standard SQL and AI. functions (preferred over dedicated tools).
golang-samber-hot
In-memory caching in Golang using samber/hot — eviction algorithms (LRU, LFU, TinyLFU, W-TinyLFU, S3FIFO, ARC, TwoQueue, SIEVE, FIFO), TTL, cache loaders, sharding, stale-while-revalidate, missing key caching, and Prometheus metrics. Apply when using or adopting samber/hot, when the codebase imports…
vector-db
Vector database expert for embeddings, similarity search, RAG patterns, and indexing strategies.