Skip to main content
prompt GP ChatGPT

phase-10l-murph-continuity-researcher — Claude Code prompt

phase-10l-murph-continuity-researcher investigates how Murph handles context compaction and interrupted turns, then produces a code-backed fix and test plan. It expects local Murph and Phantom repositories.

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-10l-murph-continuity-researcher.md
Open on GitHub

What phase-10l-murph-continuity-researcher does

This prompt directs Claude Code to inspect Murph’s context handling, compaction, provider adapters, and tool-call history, alongside Phantom transcript-recovery notes. The research should establish what the next provider request contains after compaction and whether Murph can resume an interrupted turn. It explicitly forbids editing files.

The requested output is an architecture note with file and line references, a verdict on where the fix belongs, a ranked implementation plan with acceptance criteria, and risks. It asks for provider-specific checks involving OpenAI, Anthropic, and ZAI, plus deterministic state-machine tests and real-provider tests where practical.

Findings must come from code that was actually inspected; summaries must not retain secrets or full tool payloads. The prompt targets working copies under /Users/truffle/work, so source-level verification depends on access to those repositories.

What it helps with

  • trace what Murph sends after context compaction
  • check whether Murph reuses Pi overflow and summarization primitives
  • investigate why an agent loses its answer after compaction
  • rank fixes and define acceptance criteria
  • plan tests for interrupted turns and tool-call history

Example prompts for the agent

  • Inspect how Murph resumes an interrupted turn after compaction and write an architecture note with code references.
  • Compare Murph’s context-overflow handling with Pi and recommend a production fix with acceptance criteria.
  • Review OpenAI tool-call ordering risks and propose tests for the continuation flow.

How to install phase-10l-murph-continuity-researcher

Manually: put the files in place

.claude/commands/phase-10l-murph-continuity-researcher.md

Questions about phase-10l-murph-continuity-researcher

How do I install phase-10l-murph-continuity-researcher in Claude Code?

Save the prompt as .claude/commands/phase-10l-murph-continuity-researcher.md.

What does phase-10l-murph-continuity-researcher do?

It asks Claude Code to investigate context compaction and turn continuation in Murph and Phantom. The deliverables include a code-referenced architecture note, an implementation plan, and tests.

Which files does the research cover?

It directs Claude Code to inspect Murph materials and Phantom transcript-recovery documents under /Users/truffle/work. It also suggests checking local pi-mono and pi-code copies.

Source text

phase-10l-murph-continuity-researcher.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 Phantom on Murph
post-compaction continuity. Do not edit files. Do not summarize from memory.
Read the actual Murph and Phantom code.

Context:

- Phantom is being moved onto Murph, a clean-room pi-mono based agent runtime.
- We already implemented Phantom transcript recovery through
  `phantom_chat_transcript_search`.
- Real browser testing showed long-running tool-heavy tasks work, and Murph
  compaction appears to fire, but after compaction the agent may lose the
  created-page answer or stop responding usefully.
- The operator believes compaction should be an LLM-written compact summary,
  not literal history replay. The next model call should receive that summary
  plus enough pointers and tools to continue.
- We must reuse pi-mono/Pi capabilities where they exist and avoid rebuilding
  battle-tested primitives in worse form.

Reading list:

1. `/Users/truffle/work/murph/AGENTS.md`
2. `/Users/truffle/work/murph/QUALITY-BAR.md`
3. `/Users/truffle/work/murph/PROGRESS.md`
4. `/Users/truffle/work/murph/src/**` areas related to context overflow,
   compaction, provider adapters, message history, tool call preservation, and
   summaries.
5. `/Users/truffle/work/phantom-murph-hardening/research/chat-experience/phase-10g-pi-continuity.md`
6. `/Users/truffle/work/phantom-murph-hardening/research/chat-experience/phase-10i-transcript-recovery-implementation.md`
7. Any local pi-mono, pi-code, or notes repo already cloned under
   `/Users/truffle/work` that contains relevant compaction or overflow logic.

Questions to answer:

1. What exactly does Murph do today when context approaches or exceeds the
   provider limit?
2. Does Murph use pi-mono/Pi primitives for context overflow detection,
   summarization, and retry, or did we duplicate that logic?
3. After compaction, what is sent to the next provider call? Include the shape
   of summary, preserved tool-call state, and any continuation instructions.
4. Is there an existing Murph seam for "continue the interrupted turn after
   compaction", or is compaction only a pre-call trimming step?
5. Which provider-specific constraints matter for OpenAI, Anthropic, and ZAI
   around tool-call history after compaction?
6. What is the smallest production-grade fix that improves continuation
   without corrupting tool-call protocols?
7. What tests should prove this, preferably with real providers when possible
   and deterministic unit tests for the core state machine?

Deliverables:

- A concise architecture note with file references and line numbers.
- A clear verdict: reuse Pi/Murph existing primitives, fix Murph primitive, or
  add Phantom-specific continuation glue.
- A ranked implementation plan with acceptance criteria.
- Explicit risks and anti-patterns, especially anything that would break
  OpenAI tool-call message ordering.

Anti-patterns:

- Do not suggest raw transcript stuffing.
- Do not suggest regex-heavy summarization.
- Do not expose private chain-of-thought.
- Do not preserve raw secrets, magic links, auth headers, or full tool payloads
  in summaries.
- Do not treat a single reference repo as authoritative.

Self-review:

Before finalizing, re-read your own findings and remove any claim not backed by
files you actually inspected.

© ghostwright · Apache-2.0 · ghostwright/phantom

Similar prompts

All prompts for Agent setup: memory, context, orchestration →
gsd-phase-researcher davepoon gsd-phase-researcher researches a phase before planning and writes findings to RESEARCH.md for gsd-planner. Start it through /gsd:plan-phase or /gsd:research-phase. CX Codex ★ 3,785repo stars cost-compact-context ruvnet cost-compact-context — a skill that wraps getTokenOptimizer().getCompactContext() for token cost analysis. Retrieves compacted context from ReasoningBank, reports tokens saved via bridge measurements, and exposes agent-booster availability. CX Codex +1 728installs gsd-ai-integration-phase open-gsd gsd-ai-integration-phase — a skill for Claude Code that generates an AI-SPEC.md design contract for AI system development phases. It triggers a research and evaluation planning workflow when executed. CX Codex +1 571installs Phantom Connect SDK phantom Phantom Connect SDK — an MCP server with one search tool over the Phantom developer docs. It returns code samples, API references and page links while you build wallet integrations. CX Codex +1 3,016installs revision-continuity danjdewhurst revision-continuity revises existing story chapters and catches continuity errors without losing track of character state. Built for developmental edits, timeline audits, and scene cut decisions. CX Codex +1 1,000installs series-continuity danjdewhurst The series-continuity skill lets you write sequels, prequels and spin-offs by linking multiple Story Skills projects into a series. It tracks chronology, syncs characters and locations across books, and validates canon consistency. CX Codex +1 525installs

From the same set38

The whole set →
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 phase-10l-phantom-continuity-researcher phase-10l-phantom-continuity-researcher is a Claude Code prompt for diagnosing session continuity in Phantom, including context compaction, transcript recovery, and SQLite state analysis. GP ChatGPT ★ 1,497repo stars phase-10j-tool-card-collapse-research phase-10j-tool-card-collapse-research is a prompt for Claude Code and ChatGPT that examines Phantom tool cards: collapsing completed calls while keeping errors and blocked work visible. GP ChatGPT ★ 1,497repo stars phase-10i-transcript-search-research phase-10i-transcript-search-research guides an architectural study for historical chat transcript search in Phantom after Murph compaction. Useful when planning agent context retrieval. GP ChatGPT ★ 1,497repo stars phase-10a-activity-correctness-reviewer The phase-10a-activity-correctness-reviewer prompt asks Claude Code to review Phantom’s Phase 10A patch and report concrete findings with file and line references. It targets changes in the named project. GP ChatGPT ★ 1,497repo stars phase-10m-chat-ui-polish-review phase-10m-chat-ui-polish-review — a read-only review prompt for Phantom's Murph chat UI: it walks the components, the run timeline and artifact cards, then ranks what to fix next. GP ChatGPT ★ 1,497repo stars

Project sponsors

Companies that keep this analytics open