⚖MikeWatch Forks of the Mike legal-AI codebase, distilled for legal-tech readers
Home · emileriksenkeev/sydOS · this thread

emileriksenkeev turns sydOS into a litigation case-assessment desk

This fork combines litigation-specific review templates with a chat assistant that can navigate the case record by its structured index.

workflowdiscovery

The new litigation workflows are built around the practical question every case team faces early: what is in the record, what supports each side, and where are the weak spots?

  • A case document index now records where each item came from and who authored it.
  • A litigation assessment guides teams through the case snapshot, claims and defences, evidence against legal elements, risks, strengths and vulnerabilities.
  • The project chat can consult tabular review grids, using them as an index to the record rather than relying on document names alone.
  • Re-running a review now preserves earlier good results if an interrupted run fails, and keeps columns consistent over time.
So what Litigators and legal-ops teams who want a more disciplined first-pass view of a matter should look closely at this fork.

View this fork on GitHub →

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

5 commits from emileriksenkeev/sydOS, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
8e6ee7ee Syd OS: add 3 Litigation tabular review workflows emileriksenkeev 2026-07-11 ↗ GitHub
commit body 
Grounded in the Ora/IPOPHL trademark matter document set (pleadings,
affidavits, orders, exhibits). Adds to the Built-in library (practice
"Litigation"):
- Case Document Index - master index across a case file (type/date/party/
  summary/relief/deadlines/parties/status)
- Judicial Affidavit & Witness Review - affiant/facts/exhibits/admissions/
  cross-exam points
- Orders, Rulings & Deadlines Tracker - disposition/obligations/deadlines/
  compliance status

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
098fdd26 Syd OS: enrich Case Document Index - add Source/Origin (work product vs court vs opposing vs exhibit) + explicit Author emileriksenkeev 2026-07-11 ↗ GitHub 
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
3ee0eb6e Syd OS: add flagship Litigation Case Assessment (PH) assistant workflow emileriksenkeev 2026-07-11 ↗ GitHub
commit body 
Methodology-driven, risk-rated case assessment (snapshot, claims/defenses,
elements & evidence mapping, evidence map, procedural posture & deadlines,
risk table, strengths/vulnerabilities, next steps). Added to both builtin
registries so it shows in the Built-in tab and is runnable by the assistant.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
6f63ca9b Syd OS: let the project assistant read tabular reviews as an index emileriksenkeev 2026-07-11 ↗ GitHub
commit body 
Adds list_tabular_reviews + read_tabular_review tools (project-scoped, via
db+projectId) to the project chat, plus system-prompt guidance to consult a
tabular grid (e.g. the Case Document Index) to locate/relate documents before
reading full texts. Previously read_table_cells was siloed to the tabular
chat, so the project assistant navigated by filename only.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
6c7f015d Tabular review: stop losing cell values on re-runs; make runs converge emileriksenkeev 2026-07-17 ↗ GitHub
commit body 
Four compounding reliability bugs in /tabular-review/:id/generate:

- Cells queued for processing had their stored content nulled before the
  new value existed, so any model failure (or an interrupted run) left a
  previously-good cell permanently blank. Keep the old content; when a
  column never comes back, restore the prior value instead of erroring.
- Every document ran concurrently via unbounded Promise.all - ~100
  simultaneous LLM streams on a Vault-sized review tripped provider rate
  limits, failing a different random subset of rows on every run. Use a
  bounded worker pool (TABULAR_DOC_CONCURRENCY, default 4).
- Columns the model omitted from its single streamed response were
  immediately marked error. Retry the missing columns once per document
  before giving up.
- Streamed JSON lines wrapped in markdown fences were dropped silently;
  skip fence lines instead.

Re-running a review now only fills gaps and converges instead of
churning values.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-908.md from inside the repo you want the changes in.

⬇ Download capture-thread-908.md