Peanut Butter & Chocolate v2.0: Less Plugin, Same Pipeline

Peanut Butter & Chocolate v2.0

In February, I released Peanut Butter & Chocolate, my Claude Code plugin for structured agentic coding.

The core pipeline has not changed.

Research. Plan. Implement. Review.

That is still the loop. What changed is everything around it.

PBC v2.0 is simpler. It has fewer files. It trusts the model more. It also has a Codex version now.

OpenCode Is Gone, Codex CLI Is In

The first version used OpenCode to talk to Codex. That made sense at the time. OpenCode was my daily coding agent, and it gave me a clean way to hand implementation packets to Codex.

PBC v2.0 uses the Codex CLI directly.

codex exec -m gpt-5.5 -c model_reasoning_effort="low" -c service_tier="fast" < "$PACKET_PATH"

This does not remove a dependency. It swaps OpenCode for the Codex CLI, which is the tool doing the implementation work anyway.

If you have the Codex CLI authenticated and you have access to gpt-5.5, you are good.

Commands Are Now Skills

Claude Code deprecated commands/ in favor of skills.

That is the right direction. Commands were a reasonable starting point, but skills fit the workflow better. They carry instructions, context, and invocation behavior in one place.

PBC now has seven skills instead of five commands.

Skill	Purpose
/pbc:research-codebase	Parallel sub-agent codebase exploration
/pbc:create-plan	Interactive planning with optional stress testing
/pbc:grill-me	Decision-tree interrogation of a plan or design
/pbc:implement-plan	Packet-based Codex implementation
/pbc:review-work	Independent review loops with triage
/pbc:address-pr-comments	PR comment triage and fix delegation
/pbc:handoff	Session continuation documents

The shape is the same. The packaging is cleaner.

Named Agents Are Gone

The first version shipped eight named Sonnet sub-agent definitions. codebase-locator, codebase-analyzer, pattern-finder, and several more.

I removed all of them.

The skills now describe what needs to be researched. Opus decides how to delegate using Claude Code's Agent tool. There are no separate agent spec files to maintain.

The research quality has not dropped because the useful guidance was never the agent name. It was the instruction.

The named agents were adding indirection. They made the plugin feel more deterministic, but they also made it more brittle.

grill-me

/pbc:grill-me is new.

This one is inspired by Matt Pocock. The idea is simple. Before implementation starts, stress-test the plan.

It walks the decision tree. Scope. Data model. APIs. Migration. Testing. Rollout. Security. Failure modes.

It asks one question at a time with AskUserQuestion, so each answer can shape the next question. It does not dump twenty questions on you and pretend that is collaboration.

It also checks whether it can answer the question itself before bothering you. It explores the codebase. It can use WebSearch when that is appropriate. If the answer is already knowable, it should not ask.

You only get asked questions that require human judgment.

grill-me can run standalone against a plan or design. It can also run inside /pbc:create-plan before the final plan is written.

Is it perfect? No. But it catches unclear decisions before Codex starts writing code, and that is exactly where I want those problems found.

handoff

/pbc:handoff is also new.

This is also inspired by Matt Pocock. Long sessions end. Context gets compacted. Humans go to sleep. Work still needs a clean continuation path.

The handoff skill writes a continuation document to thoughts/handoffs/. It references artifacts by path instead of duplicating them. It includes branch state, commit state, what happened, what is still open, and which skills to run next.

Most of the time, the artifact chain is already the handoff mechanism.

Research finishes and suggests /pbc:create-plan with the research artifact path. Create-plan finishes and suggests /pbc:implement-plan with the plan path. Implement-plan produces packets. Review-work produces review artifacts and fix packets.

The explicit handoff skill is for session boundaries. It is the thing I run when I want a fresh session to pick up without guessing.

Less Complex, Same Spirit

PBC v1 had 30 files. PBC v2.0 has 11.

That is the main change.

The pipeline is still Research, Plan, Implement, Review. Independent review loops still catch bugs. The artifact trail still gives traceability. Human gates still exist at every step.

The difference is that the plugin trusts the model more.

Instead of prescribing which sub-agent to spawn with a detailed spec, the skill says what to research and lets Opus figure out the details. That is less deterministic. It is also less brittle.

I care more about the artifact quality than the ceremony that produced it.

address-pr-comments Is One Of My Favorite Skills

/pbc:address-pr-comments is one of my favorite skills.

Even after an independent review pass, you still get PR comments. Humans see things the agents miss. AI reviewers leave useful notes too. The annoying part is turning that pile of comments into clean follow-up work.

This skill fetches the comments and triages each one.

Fix now
Defer
Skip

It presents the triage for approval before anything changes. Then it generates one fix packet per comment.

One packet per comment gives clean traceability. The user alignment gate means I am never surprised by what the agent decided to fix.

It is a quick way to get review feedback sorted, approved, and resolved without losing the thread.

There Is Now A Codex Version

PBC started as a Claude Code plugin, but the workflow is not Claude-specific.

There is now a Codex plugin at github.com/mattlgroff/groffdev-codex-plugin with the same skills at feature parity.

The Codex version spawns codex exec sub-agents instead of using Claude Code's Agent tool. The workflow and artifact structure are identical.

That is the part I care about. Research artifacts. Plan artifacts. Implementation packets. Review loops. Human gates.

The tool can change. The workflow should survive that.

Installation

For Claude Code:

claude plugin marketplace add mattlgroff/pbc
claude plugin install pbc@mattlgroff-pbc --scope project

For Codex, use the README in the Codex plugin repo:

open https://github.com/mattlgroff/groffdev-codex-plugin

Both versions require the Codex CLI with gpt-5.5 access.