vs local agents

Forge vs Claude Code & Codex

A local coding agent is the best thing that has happened to a developer's keyboard — and both vendors' own documentation says the same thing about it: you are the reviewer. Forge puts a named team around that agent, with a manager who judges every handoff and a record of why, so the work can run when nobody is sitting there.

Their docs already tell you who the reviewer is

Codex's own security page says to treat its output "like any other PR: run targeted verification, review diffs, and document decisions." Claude Code asks permission before it acts and puts you in the orchestrator's chair. That is honest advice and it is correct — it also means the last line of defence is a person who has to be there. In Forge that reviewer is a role in the system, and it gets counted.

A team of named people, not one capable agent

Forge does not run five copies of a general assistant. It runs a product manager, an architect, a developer, a tester and a reviewer — each a persona with a name, a job description, its own memory and its own model. Handoffs are commits authored by that persona, so you get to know who tends to over-scope and who catches the edge cases.

Autonomy means it runs when you are asleep

Both tools can now work for hours, and both are attached to a session you started and supervise. A Forge run starts from a webhook, a schedule or a chat message, executes on your own runners, escalates to a person only when its manager says a person is needed, and reports back to Telegram or Slack when it is done.

The argument is committed, not just the answer

A terminal session leaves a diff and a scrollback you will close. A Forge run leaves a branch holding the spec, the original brief and every role's handoff as separate commits. You can diff a second attempt against the first, read the verdict that sent work back, or correct a handoff yourself mid-run.

We are not routing around these agents — we run on them

A Forge role turn is executed by exactly this class of coding agent, on your hardware, through a provider pool you control. This page is not an argument that local agents are bad. It is an argument about the layer above them, which today is you.

What these tools are, in their own words

Claude Code invites you to “Hand Claude a bug fix, test, or multi-day migration” and to “Steer and review from your terminal, IDE, Slack, or web.” It “builds the plan, asks clarifying questions, and handles work that runs for hours or days”, and it “asks for permission before making changes to your files or running commands.” Since August 2026 it “runs in auto mode by default on Pro, Max, and Team plans, so it can work longer while still catching risky commands”, and Projects let you “run and easily supervise multiple Claude agents at once.” (Source: claude.com/product/claude-code .)

Codex runs in your terminal, your IDE and the cloud, with an OS-enforced sandbox and an explicit approval policy — “When Codex must ask you before it executes an action.” By default “the agent runs with network access turned off”, and it “asks for approval to edit files outside the workspace or to run commands that require network access.” Its full-access mode is labelled “Elevated Risk” and "(not recommended)." (Source: developers.openai.com/codex .)

Both are extremely good at what they do. If you are writing code with your hands on the keyboard, neither this page nor Forge is trying to take that away from you.

The sentence that this whole page is about

It is in Codex’s own security documentation, and it is the correct advice:

“Treat Codex suggestions like any other PR: run targeted verification, review diffs, and document decisions.”

Read it again as an operations statement rather than a safety note. Verification, review and documentation are assigned to you. Claude Code says the same thing in a different register: it asks permission before it acts, and it describes you as the architect and orchestrator. Notion’s co-founder, quoted on Claude Code’s own page, puts the consequence plainly:

“A big part of my job now is to keep as many instances of Claude Code busy as possible.”

That is a real and productive way to work. It is also a description of a job that does not scale past one attentive person, and that stops entirely when that person closes the laptop.

The difference in one table

Claude Code / CodexForge
What an agent isOne highly capable general agent, running your taskNamed roles with job descriptions — PM, architect, developer, tester, reviewer, researcher, writer — each a persona with its own memory
Who judges the workYou, at the approval prompt and at review timeA management agent on every handoff: pass, send back with rework instructions, or escalate to a person
What starts the workYou, in a session you are attendingA webhook, a schedule, a chat message, or a labelled issue
Where it runsThe machine you are sitting at (or the vendor’s cloud)Your own self-hosted runners, or managed — not tied to your laptop being open
What is left afterwardsA diff, and a scrollback you will closeA branch: spec, brief and every role’s handoff as commits, authored by the persona
A rejected stepYou re-prompt, and the earlier attempt is goneA new commit; the first attempt stays in history and git diff shows what changed
Escalation to a humanEvery risky command, to whoever is at the terminalOnly when the management agent judges a person is needed — to Telegram or Slack, with the decision attached
ScopeSoftwareSoftware delivery, plus research, content and operations on the same engine

Where Forge is different

It is a team of people, not an agent

This is the part that reads as marketing until you look at a run. A Forge delivery run is not five copies of one assistant with different instructions pasted on top. It is five characters: a product manager who surfaces ambiguity instead of inventing an answer, an architect, a developer, a tester and a reviewer. Each has a name, a written job description, a stat sheet, accumulated experience and its own memory of previous work in your organisation. See Character Agents and Custom Roles .

The practical effect is not flavour. Because the roles are distinct people with distinct jobs, the handoff between them is a real event that can be inspected, judged and rejected — which is exactly what does not exist when one agent does everything in one context window. And because each handoff commit is authored by the persona, you come to know them: which one tends to over-scope, which one catches the edge case, which one keeps asking the question everyone else skipped.

Somebody independent reads the work before it moves on

Between every pair of roles sits a management agent that reads the handoff and decides: pass it on, send it back with specific rework instructions, or escalate to a person. It is not the agent that produced the work grading itself, and it is not you at 11pm. See Multi-Agent Pipeline .

This is the mechanism a single agent structurally cannot have. An agent working alone can double-check itself — and both of these tools do — but the checker and the author share a context, a plan and a set of assumptions. The whole point of a reviewer is that they do not.

Autonomy is what happens while nobody is watching

Both tools now run for hours. The difference is what starts the work and who is present. A Forge run begins because an issue was labelled, a schedule fired, a message arrived in Telegram, or another run finished — not because somebody opened a terminal. It executes on runners you own, inside credentials scoped to the run, and comes back with a pull request and a report. See Autonomous Scheduled Workflows and Isolated Stacks .

When it genuinely needs a person, it says so and waits — in Telegram or Slack, with the specific decision attached — rather than blocking on an approval prompt in a terminal nobody is looking at. See Telegram Control Plane .

The reasoning is in your repository

Every run gets a branch — forge/workflow/{workflowId}/run/{runId} — holding the spec, the original brief, and each persona’s handoff as its own commit. Six months later you can read why a decision was made, not only what shipped. You can diff a second attempt against the first, or intervene mid-run by committing a corrected handoff yourself. See Git-Backed Audit .

It is not only software

The same engine, the same management agent and the same audit trail run a research team that verifies its own sources, a content team that publishes weekly, or an operations rotation. Software delivery is the most mature shape because it is how Forge builds itself — but it is one shape. See the jobs teams hand over .

The number a single agent cannot report

A local agent can tell you how many suggestions you accepted. It cannot tell you how often the work was sent back, because the reviewer is you and nobody is counting. Here is what it looks like when the reviewer is part of the system.

Twelve weeks of Forge building Forge
Runs dispatched322
Handoffs judged by a management agent1216
Sent back to their author for rework56
Escalated to a human24
Runs where the supervisor stopped at least one handoff51
Runs that completed215
Runs that failed, escalated or were cancelled103
Median wall-clock, dispatch to finished run39 minutes
Median handoffs per completed run5
Pull requests merged, across 7 repositories202

Where these came from. One workflow — Main Forge SDLC, the one that builds Forge — every run it made in twelve consecutive weeks, counted from the control plane’s own records. Not a sample and not a pilot.

Forge stopped its own work 80 times. 56 handoffs went back to their author with specific rework instructions; 24 went to a person. One run in six was interrupted by its own supervisor before any human looked at it. That is the review both vendors’ documentation correctly assigns to you — here it is priced at one model call per edge, and counted.

103 runs did not complete. We are showing you that number on the same page as the good one, because a team you can trust is one that tells you when it failed.

What we are not claiming

We are not claiming Forge writes better code. A Forge role turn is executed by exactly this class of coding agent — the difference is not the model, it is the structure around it.

We are not claiming you should stop using them. For work you are doing right now, with your hands on the keyboard and your judgement in the loop, a local agent is the better tool and Forge does not replace it. Most of our own engineers have one open.

The claim is narrower and, we think, the one that matters: the moment the work should continue without you sitting there, somebody has to hold the standard. Today that somebody is you, in a terminal. Forge makes it a role — an independent one, on every handoff, with the verdict written down.

Sources

You already run a local agent. Who reviews it when you are not there?

Point Forge at one repository and one well-specified ticket, and read the handoff commits. That is the part a terminal cannot show you.