Who it is forA developer or automation owner who manages repositories from scripts or works in an agent other than Codex.
The problemCodex is not always available, but the project still needs the same memory files and runtime.
The outcomeThe target project contains the protocol, portable runtime, and adapters, ready to be committed.
What you need first
- Codex in the ChatGPT desktop app or Codex CLI
- Python 3.9 or newer available as python3
- Git, when installing from GitHub
Steps
- Run
git clone https://github.com/psinetron/echoes-vault-codex.gitonce. - Run
python3 echoes-vault-codex/scripts/echoes_vault.py --workspace /path/to/your/project --agent codex --adapter-version 1.1.1 init, replacing the workspace path with your project. (inferred, check the README) - Open the target project and check that EchoesVault/ and .echoes-vault/ were created.
- Commit the generated files in the target project.
Prompt for your agent
Clone psinetron/echoes-vault-codex and run its scripts/echoes_vault.py init against <project-path> with --agent codex and --adapter-version 1.1.1. Expected output: the EchoesVault/ and .echoes-vault/ folders in that project. Verify that .echoes-vault/echoes_vault.py exists in the target project.
Caveat. The README says the generated runtime is the execution source after bootstrap, so the bootstrap step is only needed once per project.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for psinetron/echoes-vault-codex and "Bootstrap a project's memory from the terminal"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer