Every AI session transcript behind Eleven, published whole, secrets redacted. If an AI vendor can see it, you should be able to see it too.

The policy β€” what's published, what's redacted, what can be withheld

πŸ€– Eleven is largely built with AI, and this is the provenance record. Each entry below is a complete Claude Code session transcript β€” the prompts, the model's replies, and the commands it ran. Credentials are the only thing removed: every secret, token, key, and the private ops runbook are redacted in place, and every redaction is annotated with why (legend below).

πŸ€– Today this covers Paul's own sessions, published by his choice. The intent is that every contribution built with AI includes its prompt transcript β€” that becomes a requirement for all contributors only once agreed with them, not by surprise.

πŸ€– Contributors also keep a small personal redaction budget: a handful of things anyone may remove for personal reasons, no justification owed. Those show as redacted: personal and are counted in the open like every other redaction β€” the published count is what keeps the budget honest.

πŸ€– Sessions are in the ledger by default β€” opting in is what contributing means here β€” but a whole session can be withheld (it stays on the record below: date, prompt count, and a reason, with the content withheld) or, in the rarest cases, omitted with only the aggregate count published. Withholding is for what redaction can't fix: where sharing would cause or attract harm to a person, would reveal secrets whose exposure could cause harm (including national security), or would carry consequences beyond revealing the human's intent β€” the thing transparency is for. One case is mandatory, not optional: any session whose content could compromise the security of the centralised services (the push relay, home, cloud and billing, connect, the router) is withheld whole, because a credential can be redacted but a narrative that maps the defences cannot.

What gets redacted, and why

πŸ€– Verifying the record: each contributor's corpus carries a manifest.sha256.json naming the hash of every published file, and its history in the public releases repo's hash ledger is the tamper-evidence β€” if a published transcript ever changed after the fact, that history shows when.

πŸ€– Prefer the story to the log? Milestones β€” the key moments of building Eleven, each linked to the session where it happened.

Sessions

Local transcripts prune after roughly a week, so the registry accrues from each run of tools/prompt-registry.mjs forward; sessions that pruned before the first run are lost. Newest first. Search matches summaries and every prompt inside each session.

WhoModelSummary