Big-AGI is an open-source workspace for using multiple AI models through chat and other AI functions. It is intended for engineers, founders, researchers, and other users who want to work with AI personas, model comparisons, image generation, voice, documents, and code-related features. The catalogue entries provide commands, instructions, and a skill for working with Big-AGI.
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 commands/enricoros/big-agi/update-models-dynamicgit clone --depth 1 https://github.com/enricoros/big-AGIWrote 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/commands/enricoros/big-agi/update-models-dynamic)<a href="https://agentmods.dev/commands/enricoros/big-agi/update-models-dynamic"><img src="https://agentmods.dev/badge/commands/enricoros/big-agi/update-models-dynamic.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00043 | $0.01820 |
| Opus 5 | $0.00022 | $0.00910 |
| Sonnet 5 | $0.00009 | $0.00364 |
| Haiku 4.5 | $0.00004 | $0.00182 |
Grade A, and why
update-models-dynamic 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 5d 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Validate that the dynamic (API-fetched) vendor model parsers are up to date and not silently broken.
These vendors do NOT have hardcoded model lists - they fetch models from APIs at runtime. But their parsers, filters, heuristic detection, and capability mapping can break if upstream APIs change. This skill covers all dynamic vendors NOT covered by the other llms:update-models-{vendor} skills.
Vendors to Validate
High Risk
OpenRouter - src/modules/llms/server/openai/models/openrouter.models.ts
- Most complex parser. Vendor-specific parameter inheritance (Anthropic thinking variants, Gemini thinking/image, OpenAI reasoning effort, xAI/DeepSeek reasoning).
orModelFamilyOrderdoubles as the visibility allow-list - check if new leading vendors are missing (they'd be hidden, not just mis-sorted).orOldModelIDshiding list - check if stale.- Ids:
~vendor/model-latestmust be resolved throughalias_target.slug(dropping the '~' leaves an unlookupable ref like 'claude-opus-latest') andvendor/model-fastare resold priority tiers - both must match their base family/vendor definition, not fall through to the generic branch. reasoning.mandatorymodels reject effort 'none' - never offer it, in any vendor branch.- OR outlives vendor shutdowns: Azure serves
openai/*codex ids that are dead on OpenAI direct. Vendor-side defs must stay (hidden + deny-listed natively) for thellmOrt*Lookupinheritance to keep working - flag any OR id whose vendor lookup went dead. - Cache pricing detection (Anthropic-style vs OpenAI-style) - verify format still valid.
pricing.overrides= long-context surcharge tiers (ascendingmin_prompt_tokens, ~57/399 models) - must fold into our{ upTo, price }[]arrays, else long prompts are costed at the cheapest tier.- Variant injection for Anthropic thinking/non-thinking - verify still correct.
- Reference: https://openrouter.ai/docs/models ; live list:
GET https://openrouter.ai/api/v1/models
Medium Risk
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.
- 5d ago First seen · 102 lines · 43 tokens per session scan A 54588413b2e7
update-models-dynamic is a command published in the GitHub repository enricoros/big-AGI (7,109 stars, last pushed yesterday), licensed MIT. It adds 43 tokens to every session and 1,820 once invoked, about $0.0002 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 2026-08-30.
Other commands, from other repositories
bifrostpublicstaticblogn8n-helicone-node
🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓.
prompt
Command "prompt" from treylom/prompt-engineering-skills, covering /prompt - ai 프롬프트 생성기, ⛔ critical rules (최상단 배치), 🏛️ 대원칙 불변 조항 (재경님 2026-07-29 — 전 목적·전 모델·향후 어떤 이식/개편에도 불변), 절대 금지 사항 (constraints) and 실행 트리거 (only these).
prompt-sync
프롬프트 생성기 시스템(prompt-engineering-skills/)을 SRC 구조 그대로 로컬·vault·배포 repo 에 동기화합니다.
prompt-update
프롬프트 엔지니어링 통합 업데이트 (tofu-at 생성 — /prompt-update).
auto-prompt
AI 프롬프트 자동 생성 - K-AI 스테이션 양식 호환 (한 행에 한 모델).
music-suno-prompt
Grounded Suno prompt synthesis from local knowledge corpus + persona canon + label canon. No vibes-prompting.