Skip to main content
prompt GP ChatGPT

phase-10h-pi-thinking-research — Claude Code prompt

The phase-10h-pi-thinking-research prompt asks Claude Code to examine how Pi presents model thinking, tool activity, and long-running work, then write chat UI guidance for Phantom.

Apache-2.0 ★ 1,497 updated 2026-06-16

Install into Claude Code · Manually: put the files in place All methods and platforms ↓
.claude/commands/phase-10h-pi-thinking-research.md
Open on GitHub

What phase-10h-pi-thinking-research does

Claude Code is instructed to inspect local guidance and the Murph, Phantom, and pi-mono sources, consulting primary sources when needed. The requested report covers how Pi represents model thinking, streamed text, tool calls, progress, usage, and task completion. It also calls for Phantom chat recommendations, acceptance criteria, and risks involving provider protocols and unsafe content.

The prompt specifies a reading order and asks for file paths or primary-source links to support factual claims. It rules out edits to Phantom and Murph, as well as commits, pushes, and pull requests. Access to the listed local files is essential; without it, Claude Code cannot follow the requested research plan.

The report belongs at phase-10h-pi-thinking-research.md at the full path given in the prompt. This is a research and architecture task, not an instruction to implement the proposed UI changes.

What it helps with

  • identify which progress-display primitives Pi already provides
  • compare Pi patterns for CLI and web rendering
  • decide which chat UI responsibilities belong to Phantom, Murph, or Pi
  • write acceptance criteria for the next UI-building slice
  • review provider-protocol and unsafe-content risks

Example prompts for the agent

  • Research how Pi displays thinking and tool progress, then write the report to the path in this prompt.
  • Compare Pi's CLI and web rendering patterns and recommend an approach for Phantom's chat UI.
  • Review the research report's factual support and acceptance criteria.

How to install phase-10h-pi-thinking-research

Manually: put the files in place

.claude/commands/phase-10h-pi-thinking-research.md

Questions about phase-10h-pi-thinking-research

How do I install phase-10h-pi-thinking-research in Claude Code?

Add .claude/commands/phase-10h-pi-thinking-research.md to your Claude Code project.

What does phase-10h-pi-thinking-research do?

It directs Claude Code to research Pi's handling of thinking and activity, then produce architecture guidance for Phantom's chat experience.

Which files does the prompt need?

It requires access to the local guidance and source files under /Users/truffle/work/murph, /Users/truffle/work/phantom-murph-hardening, and /Users/truffle/work/pi-mono. Without those paths, the requested research plan cannot be carried out.

Source text

phase-10h-pi-thinking-research.md · Apache-2.0
ultrathink. ultrathink. ultrathink.

You are a principal engineer, principal architect, and principal product
manager at Anthropic, all three at once. You are researching how Phantom should
render agent thinking, tool progress, and long-running activity now that
Phantom runs on Murph, which runs on pi-mono.

Mission:

Find what Pi and Pi-adjacent code already provide for thinking display,
progress display, activity rows, run timelines, and CLI or web rendering. Do
not let Phantom reinvent primitives Pi already gives us. If Pi does not provide
the UI primitive, identify the cleanest Phantom-owned layer.

Required reading, in order:

1. `/Users/truffle/.claude/AGENTS.md`
2. `/Users/truffle/work/murph/QUALITY-BAR.md`
3. `/Users/truffle/work/murph/PROGRESS.md`
4. `/Users/truffle/work/phantom-murph-hardening/research/chat-experience/phase-10g-pi-continuity.md`
5. Local pi-mono sources under `/Users/truffle/work/pi-mono`, especially:
   - `packages/agent/src/`
   - `packages/ai/src/types.ts`
   - `packages/ai/src/providers/`
   - `packages/web-ui/README.md`
   - `packages/web-ui/src/`
   - `packages/mom/src/`

External source rules:

- Use primary sources only. GitHub repositories, official docs, package
  READMEs, and source code are acceptable.
- If you need to clone another Pi/Pi Code related repository from pi.dev or
  GitHub, clone it under `/Users/truffle/work/research-clones/`.
- Do not use SEO blogs, scraped docs, or tutorials as evidence.
- Do not copy implementation from any repo into Phantom. This is research and
  architecture guidance only.

Deliverable:

Write `/Users/truffle/work/phantom-murph-hardening/research/chat-experience/phase-10h-pi-thinking-research.md`.

The report must include:

1. Source inventory: exact repos/files read and why they matter.
2. How Pi represents thinking, streaming text, tool calls, tool progress,
   usage, and completion.
3. Whether Pi has a CLI or web rendering pattern we should reuse directly,
   adapt conceptually, or ignore.
4. What Phantom should own versus what Murph/Pi should own.
5. Specific recommendations for Phantom's chat UI, ordered by impact.
6. Risks and anti-patterns, especially places where a custom UI could break
   provider protocol or leak unsafe content.
7. Concrete acceptance criteria for the next builder slice.

Non-goals:

- Do not edit Phantom application code.
- Do not change Murph code.
- Do not commit, push, or open a PR.
- Do not create a broad open-source polish plan. Stay focused on chat
  experience and thinking/progress rendering.

Self-review:

Before finishing, re-read your report and verify every factual claim has a
local file path or primary-source URL. If a claim is an inference, label it as
an inference.

© ghostwright · Apache-2.0 · ghostwright/phantom

Similar prompts

Prompts catalog →

From the same set38

The whole set →
phase-10h-phantom-chat-review The phase-10h-phantom-chat-review prompt runs an in-depth audit of Phantom's chat UI and event pipeline after PR 109. It flags UI gaps and builds an implementation slice without modifying code. GP ChatGPT ★ 1,497repo stars list-plugins list-plugins reads settings.json and reports active Claude Code plugins with their marketplace IDs. Use it to check which plugins are currently enabled. CX Codex +1 ★ 1,497repo stars phase-10b-durable-timeline-reviewer phase-10b-durable-timeline-reviewer — a specialized prompt for reviewing Phase 10B of Phantom's chat system: validates timeline correctness, durability, idempotency, and data privacy. Reads diffs, identifies security issues and regressions. GP ChatGPT ★ 1,497repo stars phase-10h-chat-integrity-review phase-10h-chat-integrity-review reviews uncommitted git diffs for Phantom chat transcript integrity in Claude Code. It inspects attachment persistence and message rendering without modifying project code. GP ChatGPT ★ 1,497repo stars phase-10h-provider-thinking-research phase-10h-provider-thinking-research — a Claude slash command that digs through Murph, Pi and Phantom sources to find what providers really expose about thinking and reasoning events. GP ChatGPT ★ 1,497repo stars phase-10b-durable-timeline-re-reviewer The phase-10b-durable-timeline-re-reviewer prompt for Claude Code re-audits fixes in Phantom PR #106 Phase 10B, checking durable timelines for secret leaks and validating session resume states. GP ChatGPT ★ 1,497repo stars phase-10i-transcript-recovery-re-review The phase-10i-transcript-recovery-re-review prompt guides Claude Code through a second review of transcript-recovery fixes, including project file inspection, required checks, and a written report. GP ChatGPT ★ 1,497repo stars phase-10k-page-artifact-ui-builder phase-10k-page-artifact-ui-builder is a Claude Code prompt for extracting page artifacts from tool results and displaying them in a compact chat card. GP ChatGPT ★ 1,497repo stars

Project sponsors

Companies that keep this analytics open