Welcome to Project Turquoise

Turquoise Guardian

Official Handshake learn hub for this project

Turquoise Guardian is the Handshake AI learning portal for Project Turquoise. Use it for project training, tasking guidance, and anything the program publishes there. You may need to sign in with your Handshake AI account.

https://turquoise-guardian.learn.joinhandshake.com

End-to-end flow

From Guardian + blank agents to a Drive submission

Step 1 — Set up two blank-slate agents

Each session runs the agent with a throwaway config home — your own setup is untouched, and the agent starts with no MCP servers, no memory, no custom instructions, and a fresh working directory.

Option A Codex

Paste this whole block into your terminal (it launches at the end):

blank_codex() {
  local root codex_bin

  codex_bin="$(command -v codex 2>/dev/null)"
  if [ -z "$codex_bin" ] && [ -x "/Applications/ChatGPT.app/Contents/Resources/codex" ]; then
    codex_bin="/Applications/ChatGPT.app/Contents/Resources/codex"
  fi

  if [ -z "$codex_bin" ]; then
    echo "Codex CLI was not found. Install it or add it to PATH."
    return 1
  fi

  root="$(mktemp -d "${TMPDIR:-/tmp}/codex-blank.XXXXXX")" || return 1
  mkdir -p "$root/home" "$root/work"
  git -C "$root/work" init -q

  echo "Blank Codex root: $root"
  echo "Session logs will be under: $root/home/sessions/"
  echo "Work dir: $root/work"

  CODEX_HOME="$root/home" "$codex_bin" \
    -C "$root/work" \
    -m gpt-5.6-sol \
    -c 'model_reasoning_effort="high"'
}

blank_codex

Option B Claude Code

Recommended: paste this into any coding agent (Claude Code, Codex, Cursor, etc.) and let it launch a blank-slate Claude session for you.

Set up a Project Turquoise blank-slate Claude Code session on this Mac.

Requirements:
1. Create two temp directories with predictable names under $TMPDIR (or /tmp):
   - config: mktemp -d "${TMPDIR:-/tmp}/claude-blank.XXXXXX"
   - work:   mktemp -d "${TMPDIR:-/tmp}/claude-blank-work.XXXXXX"
   (These prefixes matter — later we recover session .jsonl from claude-blank.*/projects.)
2. Write settings.json into the config dir with exactly:
   {"model": "claude-fable-5", "effort": "high"}
3. Launch Claude Code from the empty work dir using:
   CLAUDE_CONFIG_DIR="" claude --strict-mcp-config --mcp-config '{"mcpServers":{}}'
4. Do not add MCP servers, plugins, memory, or custom instructions.
5. Print the config dir, work dir, and the exact launch command used. Remind me that session logs live under: <config-dir>/projects/
6. If `claude` is not on PATH, locate it or tell me how to install Claude Code.
7. Remind me to sign in with Claude Max if prompted, and to confirm the UI shows Fable 5 at high effort before I send my first task prompt.

Run the commands for me (or print them clearly if you cannot launch interactively).

cfg="$(mktemp -d "${TMPDIR:-/tmp}/claude-blank.XXXXXX")"
work="$(mktemp -d "${TMPDIR:-/tmp}/claude-blank-work.XXXXXX")"
echo '{"model": "claude-fable-5", "effort": "high"}' > "$cfg/settings.json"
echo "CLAUDE_CONFIG_DIR=$cfg"
echo "Session logs will be under: $cfg/projects/"
echo "Work dir: $work"
cd "$work" && CLAUDE_CONFIG_DIR="$cfg" claude --strict-mcp-config --mcp-config '{"mcpServers":{}}'

Before you begin — confirm setup

Check every box. If any of them doesn’t hold yet, fix your setup first.

• I have read the instructions above in full.
• Both agents are blank-slate: fresh installs with no MCP servers, custom tools, or special configurations.
• Codex is running GPT-5.6-Sol and Claude Code is running Fable 5, both at maximum / high reasoning.

Step 2 — Run the same task in both agents

The exchange protocol

1. Give the exact same initial prompt to both agents, word for word.
2. Work the problem back and forth with each agent as you naturally would.
3. Stop when solved, or after 15 back-and-forth exchanges — whichever comes first.

When you finish each session

If you still want the agent to write the .md itself, that is Path A (optional). Prefer Path B for normal token usage.

What to write up for the submission form

Save transcripts (two useful paths)

Project Turquoise wants a faithful Markdown transcript. Prefer recovering the raw session log and converting in tourqmd — it’s free of extra model tokens.

Option / Path B · recommended Raw session logs → tourqmd

After you finish each agent, recover the raw .jsonl, drop it into tourqmd, review Reader mode, then download or Submit.

Easiest — Terminal one-liner (after you finish each agent)

Project Turquoise blank-slate sessions do not write under ~/.claude or ~/.codex. They use the temp dirs from Step 1:

• Codex (blank_codex): ${TMPDIR:-/tmp}/codex-blank.*/home/sessions/**/rollout-*.jsonl
• Claude Code (blank setup): ${TMPDIR:-/tmp}/claude-blank.*/projects/**/*.jsonl

After you finish each agent, run the matching command in Terminal. It moves the newest session .jsonl onto your Desktop for tourqmd.

Claude Code — newest blank-slate session → Desktop:

mv "$(find "${TMPDIR:-/tmp}"/claude-blank.*/projects -name '*.jsonl' -type f -print0 2>/dev/null | xargs -0 ls -t 2>/dev/null | head -1)" ~/Desktop/

Codex — newest blank-slate session → Desktop:

mv "$(find "${TMPDIR:-/tmp}"/codex-blank.*/home/sessions -name 'rollout-*.jsonl' -type f -print0 2>/dev/null | xargs -0 ls -t 2>/dev/null | head -1)" ~/Desktop/

cp "$(find "${TMPDIR:-/tmp}"/claude-blank.*/projects -name '*.jsonl' -type f -print0 2>/dev/null | xargs -0 ls -t 2>/dev/null | head -1)" ~/Desktop/

cp "$(find "${TMPDIR:-/tmp}"/codex-blank.*/home/sessions -name 'rollout-*.jsonl' -type f -print0 2>/dev/null | xargs -0 ls -t 2>/dev/null | head -1)" ~/Desktop/

Or: paste this into a coding agent and let it locate and stage the files (still cheaper than asking the research agent to rewrite the full transcript).

Help me recover Project Turquoise blank-slate session logs for Codex and/or Claude Code on this Mac and stage them for tourqmd.

Do the following:
1. Find the newest relevant session files from the Step 1 blank-slate temp dirs (default for this project):
   - Codex:  ${TMPDIR:-/tmp}/codex-blank.*/home/sessions/**/rollout-*.jsonl
   - Claude: ${TMPDIR:-/tmp}/claude-blank.*/projects/**/*.jsonl
   Also check any printed CODEX_HOME / CLAUDE_CONFIG_DIR paths I still have open in a terminal.
2. List the newest few with full paths, sizes, and modified times.
3. Copy the ones I choose (or the newest for each tool) to a permanent folder on my Desktop, e.g.:
   ~/Desktop/turquoise-sessions/
   Name them clearly, e.g. codex-<taskid>.jsonl and claude-<taskid>.jsonl
4. Remind me to open https://tourqmd.traphic.dev/ , drop each .jsonl, review Reader mode, then Submit with my alias.
5. Do not use ~/.codex or ~/.claude unless the blank-slate temp dirs are empty — those are normal installs, not Project Turquoise blank sessions.
6. Do not upload anywhere yourself — only copy files locally and print the final paths.

If a path is missing or empty, explain what to re-run and how to save a transcript before closing the agent next time.

Option / Path A · optional Agent-written transcript

If you still choose Path A, send this at the end of the session:

Please save a faithful transcript of this conversation as a .md file titled "claude/codex-transcript-'first four characters of task id'.md". Do not include any reference to the model or harness used.

1. Use the “save a faithful transcript…” prompt only if you need it.
2. Ask the agent to move the .md to a permanent folder (Desktop / Documents).
3. Keep both files clearly named, e.g. codex-transcript-ab12.md and claude-transcript-ab12.md.

${TMPDIR:-/tmp}/codex-blank.*/home/sessions

find "${TMPDIR:-/tmp}"/codex-blank.*/home/sessions -name 'rollout-*.jsonl' -type f -print0 2>/dev/null | xargs -0 ls -lt 2>/dev/null | head
open "${TMPDIR:-/tmp}"

${TMPDIR:-/tmp}/claude-blank.*/projects

find "${TMPDIR:-/tmp}"/claude-blank.*/projects -name '*.jsonl' -type f 2>/dev/null | head
open "${TMPDIR:-/tmp}"

Convert with tourqmd

Clean back-and-forth transcripts for upload and rating

1. Open the converter.
2. Drop a Claude or Codex .jsonl (or use a sample to learn the UI).
3. Review in Reader mode (formatted). Switch to Source for raw Markdown.
4. Copy or Download .md for your records.

Output shape

# Transcript

**User:**
your message

**Assistant:**
agent reply and readable actions

**User:**
…

**Assistant:**
…

Submit to Google Drive

After both transcripts look good

1. Convert each session (Codex + Claude) in tourqmd.
2. Click Submit.
3. Enter your alias only (no email).
4. The file uploads as alias-1__….md, then alias-2__….md, and so on.

Prefer clear names before upload when possible (e.g. include codex / claude in the source filename). Your alias still drives the unique ID.

Task write-up (for the submission form)

Fill these two blocks carefully — raters rely on them

Block 1 — What you set out to do

Include:

• What you set out to do and why
• What area of coding the task sits in
• Required expertise to perform the task
• How long an expert would take to do it themselves

Block 2 — Initial prompt (identical for both agents)

Copy and paste the exact initial prompt you gave to both Codex and Claude Code.

Creating · Vetting · Blind rating

How your work is used in the project

Troubleshooting

Useful links

https://turquoise-guardian.learn.joinhandshake.com

https://tourqmd.vercel.app/

https://tourqmd.vercel.app/compare.html

https://tourqmd.vercel.app/instructions.html

https://tourqmd.traphic.dev/admin.html

You’re ready

Check Turquoise Guardian for project guidance → set up blank agents → run the same task twice → save both transcripts → convert in tourqmd if needed → submit with your alias.