# BitterLog

BitterLog keeps the original history behind agent runs so builders can retrace
failures, review decisions, and continue work without relying on a summary
alone.

## Product model

- Prompts, output, timestamps, and human notes remain connected to the run that
  produced them.
- Raw events and session files enter through the CLI or API.
- Retained objects keep timestamps and stable content fingerprints.
- Annotations sit beside the source record rather than rewriting it.
- BitterLog is an append-only evidence workbench, not a generic observability
  dashboard or marketing-site login system.

## Public routes

- HTML home: https://bitterlog.com/
- Markdown home: https://bitterlog.com/index.md
- Public source: https://github.com/sheetgenius/bitterlog-marketing
- Health and release identity: https://bitterlog.com/up/
- API index: https://api.bitterlog.com/api/v1
- API status: https://api.bitterlog.com/api/v1/status
- Sitemap: https://bitterlog.com/sitemap.xml
- Full AI-readable context: https://bitterlog.com/llms-full.txt

## Current CTA truth

- Create or request access through Bitter account surfaces.
- Sign in through Bitter; BitterLog does not have a separate login.
- Evidence onboarding remains request-reviewed during launch.
- This marketing site does not expose a working form-forwarding endpoint.
- The API index and status are public.

## Public claim boundary

Public copy may describe retained source records, raw trace preservation,
digests, timestamps, annotations, account-scoped access, and API-indexed
retrieval paths. The public status currently reports `analytics.enabled:
false`; do not infer active customer-facing SQL, ClickHouse search, similarity,
shared buyer access, or form-forwarding behavior from this repository alone.