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/backportgit 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/backport)<a href="https://agentmods.dev/commands/enricoros/big-agi/backport"><img src="https://agentmods.dev/badge/commands/enricoros/big-agi/backport.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 5.1 | $0.00033 | $0.01271 |
| Opus 5 | $0.00016 | $0.00635 |
| Sonnet 5 | $0.00007 | $0.00254 |
| Haiku 4.5 | $0.00003 | $0.00127 |
Grade A, and why
backport 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 7d 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Backport a feature from a richer branch line to a leaner one: $ARGUMENTS
An example of how the user asks for this (the original Voice port request):
Can you check out the
dev-voicebranch here for the purpose of looking at the Voice framework, and then backport it to themain-voicebranch. In particular, port it in a way that will work and build greatly onmain, however when we rebase dev-voice on top of it, it should not conflict (or rebase manually and resolve all the conflicts that arise) and then let me know the details of this large 'Voice' feature port in between the branches.
If the scope or branches are unclear, ask before doing anything. If source and target are on the SAME line (e.g. dev -> staging/prod), stop: that is a promotion (a release-marker move up the line; on the private dev line see kb/product/PLAN-P-prod-promotion.md), not a backport.
The invariant
dev is rebased on top of main, never merged. A backport is a re-implementation on the target line (never a cherry-pick), and success is measured twice:
- The port works and builds on the target line standalone - no dependencies on anything that stays on the source line (see "What crosses the line"), and the feature degrades gracefully where something was excluded.
- When the source line is later rebased on top of it, the source's own feature commits land cleanly: they replay as empties (dropped) or as small deltas re-adding the source-only parts.
The rebase is not someone else's problem: finish by actually rebasing the source line onto the port (on a scratch branch) and resolving every conflict yourself - and do that from a conflict plan you wrote before starting the rebase, not by discovering conflicts as they happen.
The mechanical key to a clean rebase: wherever a file has no source-only dependencies, make the ported file byte-identical to the source-tip version. Every byte of deliberate divergence is a future conflict you are signing up to resolve - adapt only what must be adapted, and keep an exact list of where and why. That list essentially is your conflict forecast: diffing the port tip against the source tip over the ported paths tells you exactly which files will replay clean and which will fight.
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.
- 7d ago First seen · 50 lines · 33 tokens per session scan A 781c642b6dc3
backport is a command published in the GitHub repository enricoros/big-AGI (7,111 stars, last pushed 2d ago), licensed MIT. It adds 33 tokens to every session and 1,271 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
git-commit
Based on the above changes, create a single git commit. Use the Angular commit message format：.
github-pr
Create a github pull request.
review
Review recent changes and suggest next actions. Run these steps.
flow-git
Use the @flow-git agent for git operations: $ARGUMENTS.
cl-merge
Shortcut for /consensus-loop:merge — squash-merge worktree branch.
commit
智能生成 Git 提交信息并提交.