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 agentmods add agents/modelstudioai/cli/model-add-removegit clone --depth 1 https://github.com/modelstudioai/cliWhat 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 | $0.00000 | $0.00808 |
| Opus 5 | $0.00000 | $0.00404 |
| Sonnet 5 | $0.00000 | $0.00162 |
| Haiku 4.5 | $0.00000 | $0.00081 |
Grade A, and why
model-add-remove 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 yesterday.
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.
What it actually says
模型上下架
触发条件
- 上线新的 Qwen / Wan / CosyVoice / 等模型
- 切换某命令的默认模型(如
bl text chat默认从 qwen3.7-max 切到 qwen3.7-plus) - 废弃旧模型
模型本身是阿里云后端在管,本仓库要做的是让 CLI 能正确调用 + 文档/AI 入口准确反映可用模型清单。
必查清单
A. 命令实现
-
packages/commands/src/commands/<group>/<action>.ts:--modelflag 的 description 里"default:"反映新默认值- 命令内部
const model = flags.model || settings.defaultXxxModel || "<default>"的 fallback 字符串 - 如果命令维护一个 supported-models 列表(如
speech/synthesize.ts:MODEL_VOICES),增删条目 - 如果不同模型有不同 endpoint / 请求体形状,确保
if (model.startsWith("xxx"))分支覆盖
- 模型如有特殊 endpoint,看
packages/core/src/client/endpoints.ts - 如果新增的是某产品入口专属能力,确认
packages/cli/src/commands.ts或其它入口 map 是否需要暴露/隐藏
B. 类型层
-
packages/core/src/types/api.ts的 request/response 类型如果跟模型相关,同步字段
C. 命令手册
- 若
--model的 description 含 default,改命令后跑pnpm --filter bailian-cli run generate:reference更新对应skills/<skill>/reference/<group>.md并提交 - 同步拥有该命令的领域 skill「When to use which command」表中的 Default model(现主要是
bailian-gen;精调相关看bailian-finetune正文示例)。hubbailian-cli已瘦身,一般不必再写领域默认模型(见 skill-change.md)
D. 用户面文档
-
README.md/README.zh.md:- Quick Start 示例如使用了具体型号,确认仍可用
- 顶部 introduction 段落如提到"Qwen-Omni"等品牌名,无需变(模型代号变化不算品牌变)
E. 测试层
- 按 cli-e2e-tests.md 维护 e2e:断言不硬编码废弃模型 ID;新模型至少一条 happy-path 集成
完成后自查
# 默认模型走通
pnpm -F bailian-cli exec tsx src/main.ts <command> --message "test"
# 显式指定新模型
pnpm -F bailian-cli exec tsx src/main.ts <command> --model <new-model> --message "test"
常见漏点
- ✗ 改了命令默认模型,但 SKILL.md frontmatter 或领域路由表 Default model 仍写老型号 → AI agent 调用时仍按老型号宣传
- ✗ 只改了
reference// flag description,忘改bailian-gen(等) SKILL 路由表 - ✗ 废弃模型时只删了代码,e2e 测试还在跑,CI 红
- ✗ 新模型 endpoint 不一致,但只改了 default,没加 endpoint 分支判断
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.
- yesterday First seen · 57 lines · 0 tokens per session scan A 584876976ab6
model-add-remove is an agent published in the GitHub repository modelstudioai/cli (320 stars, last pushed 4d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 808 tokens. 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 2026-08-30.
Other agents, from other repositories
index
Browse built-in Agent Framework capabilities for multimodal input, tools, retrieval, evaluation, security, and autonomous execution.
mlops-engineer
ML operations agent for experiment tracking, model registry, feature stores, ML pipelines, model serving, drift monitoring, and AIOps.
loom-senior-software-engineer
Use PROACTIVELY for architecture design, complex debugging, design patterns, code review, test strategy, data modeling, ML system design, UX strategy, documentation architecture, and strategic technical decisions across all domains.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
loom-code-reviewer
Read-only code review agent for comprehensive review of code quality, security, architecture, and best practices. Cannot modify files.
integrity-check
Detect adversarial content in .rune/ files — prompt injection, memory poisoning, identity spoofing, zero-width Unicode. Verdict: CLEAN/SUSPICIOUS/TAINTED.