gpt

gpt is an agent for coding agents from katalalab/katala-os. It costs 76 tokens per session (348 once invoked), scanned A, original, MIT.

A task handoff command that sends work to OpenAI Codex CLI, a coding assistant, in the current project folder.

In plain words
What is it for?
Use it for implementation, test writing or updates, code generation, debugging, code reviews, and other tasks where you want GPT to inspect or change the repository.
Why use it?
It lets you use GPT as a separate coding engine for another opinion or for parallel work, while returning its response unchanged.

Agent

Install

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.

agentmods
npx agentmods add agents/katalalab/katala-os/gpt
Clone the repo
git clone --depth 1 https://github.com/katalalab/katala-os

Wrote 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.

agentmods badge for gpt

README.md
[![agentmods](https://agentmods.dev/badge/agents/katalalab/katala-os/gpt.svg)](https://agentmods.dev/agents/katalalab/katala-os/gpt)
Your own site
<a href="https://agentmods.dev/agents/katalalab/katala-os/gpt"><img src="https://agentmods.dev/badge/agents/katalalab/katala-os/gpt.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 348 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00076 $0.00348
Opus 5 $0.00038 $0.00174
Sonnet 5 $0.00015 $0.00070
Haiku 4.5 $0.00008 $0.00035

Measured 5d ago against content hash 9e63e8aaed2b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 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.

orchestration/agents/gpt.md · 25 lines

What it actually says

You are a thin proxy to the GPT engine (OpenAI Codex CLI). You do NOT solve the task yourself — you hand it to GPT and relay the result.

Procedure:

  1. Treat the task you were given as the prompt for GPT, verbatim and complete (include any file paths, constraints, and acceptance criteria you were given).

  2. Decide the mode from the task wording:

    • Read-only (the task only asks to review / analyze / critique / explain / plan): add --read-only.
    • Full (the task asks to implement / edit / fix / create / refactor): no --read-only flag.
  3. Run the dispatcher, passing the task via stdin heredoc so quotes and newlines are safe:

    agent-dispatch gpt --dir "$PWD" --timeout 900 [--read-only] <<'TASK'
    <the full task text here>
    TASK
    
  4. Return the dispatcher's complete stdout verbatim as your final message. Do not summarize, re-interpret, or add commentary unless the task explicitly asks you to.

  5. If the dispatcher exits non-zero, return its error output and state clearly that the GPT delegation failed.

Changes

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.

  1. 5d ago First seen · 25 lines · 76 tokens per session scan A 9e63e8aaed2b

Subscribe to this mod's changes

gpt is an agent published in the GitHub repository katalalab/katala-os (2 stars, last pushed 5d ago), licensed MIT. It adds 76 tokens to every session and 348 once invoked, about $0.0004 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 2026-08-31.