The full text
The whole documents, one page each, exactly as they were sent.
Prompt capture
v2.1.235 system prompt
2 added and 1 removed, of 223 lines, about 3 words, in the prompt 17 of 27 arms receive. 3 other prompts also changed. Compared against v2.1.234.
This arm was captured with the sandbox seeded from flags-anon, the anonymous GrowthBook payload, seeded into the sandbox .claude.json. 514 flags from data/flags/snapshots/flags-20260818T185025Z.json. What a seeded capture does and does not tell you.
- Calling Agent with subagent_type: "fork" creates a fork — it inherits your full conversation context, runs in the back...
System prompt, reached 17 of 27 arms +3 −1 of 223 lines
This edit reached 17 of 27 arms: claude-3-5-haiku, claude-3-5-sonnet, claude-3-7-sonnet, claude-3-opus, claude-haiku-4-5, claude-opus-4-5, claude-opus-4-6, claude-opus-4-7, claude-sonnet-4-0, claude-sonnet-4-5, claude-sonnet-4-6, claude-sonnet-5, opusplan, sonnet, claude-sonnet-5[1m], haiku, sonnet[1m]
from line 68
# Session-specific guidance - If you need the user to run a shell command themselves (e.g., an interactive login like `gcloud auth login`), suggest they type `! <command>` in the prompt — the `!` prefix runs the command in this session so its output lands directly in the conversation. - - Calling Agent with subagent_type: "fork" creates a fork — it inherits your full conversation context, runs in the background, and keeps its tool output out of your context — so you can keep chatting with the user while it works. Reach for it when research or multi-step implementation work would otherwise fill your context with raw output you won't need again. Other subagent_type values (or omitting it) start fresh agents with no context. **If you ARE the fork** — execute directly; do not re-delegate. + - Calling Agent with subagent_type: "fork" creates a fork — it inherits your full conversation context, runs in the background, and keeps its tool output out of your context — so you can keep chatting with the user while it works. Reach for it when research or multi-step implementation work would otherwise fill your context with raw output you won't need again. Other subagent_type values start fresh agents with no context. **If you ARE the fork** — execute directly; do not re-delegate. - When the user types `/<skill-name>`, invoke it via Skill. Only use skills listed in the user-invocable skills section — don't guess. # auto memory
from line 219
# Context management When the conversation grows long, some or all of the current context is summarized; the summary, along with any remaining unsummarized context, is provided in the next context window so work can continue — you don't need to wrap up early or hand off mid-task. + +<total_tokens>15000000 tokens left</total_tokens>
System prompt, reached 10 of 27 arms +2 −0 of 77 lines
This edit reached 10 of 27 arms: (unset), claude-opus-4-0, claude-opus-4-1, claude-opus-5[1m], opus, claude-opus-5, claude-fable-5, claude-mythos-5, claude-opus-4-8, default
The same edit landed in 2 other copies of this prompt, which differ from this one only in text interpolated per model.
from line 73
Do not call the AgentTool unless the user requested it Do not use workflows or deep-research unless the user requested it + +<total_tokens>15000000 tokens left</total_tokens>