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/wiseaidotdev/autogpt/git-gptgit clone --depth 1 https://github.com/wiseaidotdev/autogptWhat 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 | $0.00625 |
| Opus 5 | $0.00000 | $0.00313 |
| Sonnet 5 | $0.00000 | $0.00125 |
| Haiku 4.5 | $0.00000 | $0.00063 |
Grade A, and why
git-gpt 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.
How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitGPT
Feature: git
GitGPT automates version control by creating atomic Git commits whenever an agent completes a task. It monitors agent status, stages the entire workspace, and commits with a structured message that traces the commit back to the original user goal and agent role.
What GitGPT Solves
In autonomous or quasi-autonomous workflows, tracking what changed and why is difficult without discipline. GitGPT enforces a clean, auditable Git history automatically, every agent action becomes a commit with a descriptive, machine-generated message.
Enabling GitGPT
cargo install autogpt --features git,gem,gpt
AUTOGPT_WORKSPACE must point to a directory that is a Git repository (or within one).
How It Works
GitGPT runs after each agent completes. It:
- Checks if the agent status is
Completed - Stages all modified files (
git add .) in the workspace - Computes a structured commit message from the task description and agent role
- Creates a signed commit using
GitGPT <[email protected]>as the author
Example Commit Message
commit cc448377fae752ba28847c873751ba1170d19fc0 (HEAD -> master)
Author: GitGPT <[email protected]>
Date: Mon Apr 7 00:25:59 2025 +0300
User Request: Project Goal: "Develop a patient management system
encompassing appointment scheduling, patient records, and billing features.",
Agent Role: "frontend", programming language: "Python", framework: "FastAPI"
Output:
- Implementing FastAPI, construct a frontend in Python for a patient
management system encompassing appointment scheduling, patient records,
and billing features.
- Step 1: Develop the user interface components for appointment scheduling.
- Step 2: Build the UI for managing and displaying patient records.
- Step 3: Create the billing interface for generating and displaying invoices.
SDK Usage
use autogpt::prelude::*;
#[tokio::main]
async fn main() {
let backend = BackendGPT::new(
"Backend Developer",
"Generate a REST API for a blog in Python using FastAPI.",
"python",
).await;
let git_agent = GitGPT::new("GitGPT", "Commit backend changes").await;
AutoGPT::default()
.with(agents![backend, git_agent])
.build()
.expect("Failed to build AutoGPT")
.run()
.await
.unwrap();
}
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 · 77 lines · 0 tokens per session scan A dc523e1a0309
git-gpt is an agent published in the GitHub repository wiseaidotdev/autogpt (115 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 625 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
agent_class
The Agent class is the core orchestrator in AgentForge. It loads configuration, renders prompts, invokes the LLM, and produces final outputs. Agents can be subclassed for custom logic.
agent_prompts
AgentForge uses YAML-based prompt templates to drive agent behaviors. All prompt files live under.
custom_agents
This is an advanced extension reference.
agents
Agents are the orchestrators in AgentForge, binding configuration, prompts, models, and storage into end-to-end AI workflows. An agent.
verify
Verifies edits conform to project AGENTS.md and suggests simplifications.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.