Skip to content

docs: fix inaccurate README/docs claims (verb count, dead links, stale what's-new, bad example flags)#243

Open
rylinjames wants to merge 1 commit into
mainfrom
docs/accuracy-fixes
Open

docs: fix inaccurate README/docs claims (verb count, dead links, stale what's-new, bad example flags)#243
rylinjames wants to merge 1 commit into
mainfrom
docs/accuracy-fixes

Conversation

@rylinjames

Copy link
Copy Markdown
Collaborator

Summary

  • Verb count fixed: README line 16 said "eleven verbs" and line 275 said "11 visible verbs" — the actual CLI has 18+ visible top-level commands (6 standalone @app.command() + 12 non-hidden app.add_typer() groups). Rephrased both to "a focused set of verbs" / "Core user-facing verbs shown above" to avoid a brittle exact count.
  • Dead reflex_context/ hyperlinks removed: Three clickable markdown links in README pointed into the private reflex_context/ vault (not in this repo): reflex_context/measured_numbers.md (×2) and reflex_context/03_experiments/2026-04-29-v07-runtime-spike.md. Each link is de-linked; surrounding prose is kept, with a note that it is in the private vault.
  • v0.12.0 "What's new" entry added: The top of the README's version history stopped at v0.11.2 (2026-05-29). Added a brief v0.12.0 (2026-06-03) entry covering the Reflex→Tether rename, the fastcrest-tether PyPI dist, tether CLI/import, and the reflex compat shim lifecycle (supported through v0.13.x, removed in v0.14.0). Mirrors CHANGELOG.md d444db8.
  • CONTRIBUTING.md quickstart fixed: The dev-setup snippet told contributors to run tether serve --policy smolvla --device cuda. The --policy flag does not exist on tether serve (which takes a positional export_dir). Replaced with tether go --model smolvla-base, the correct one-command smoke-test path.
  • Jetson example flags fixed (examples/02-deploy-smolvla-jetson.md line 65): tether export --model smolvla-base --out ./smolvla-export/ used two nonexistent flags. The real export signature has MODEL as a positional argument and --output (not --out). Fixed to tether export smolvla-base --output ./smolvla-export/.
  • THIRD_PARTY_LICENSES.md updated: Header said "Reflex VLA bundles or vendors code" — updated to "Tether bundles or vendors code". Four src/reflex/ paths describing where this project's own vendored code was lifted into were updated to src/tether/. Upstream third-party names, licenses, and the MIT license text are untouched.

Before / after (key changes)

File Before After
README.md:16 "One CLI, eleven verbs, plus a chat agent." "One CLI, a focused set of verbs, plus a chat agent."
README.md:275 "11 visible verbs." "Core user-facing verbs shown above."
README.md:14,484 [reflex_context/measured_numbers.md](...) `reflex_context/measured_numbers.md` (private vault, not in this repo)
README.md:116 [`reflex_context/03_experiments/...`](...) `reflex_context/03_experiments/...` (private vault, not in this repo)
README.md (no v0.12.0 entry) Added v0.12.0 rename bullet
CONTRIBUTING.md tether serve --policy smolvla --device cuda tether go --model smolvla-base
examples/02-deploy-smolvla-jetson.md:65 tether export --model smolvla-base --out ./smolvla-export/ tether export smolvla-base --output ./smolvla-export/
THIRD_PARTY_LICENSES.md "Reflex VLA bundles..." / src/reflex/ paths "Tether bundles..." / src/tether/ paths

Test plan

  • Verify README renders correctly in GitHub preview (no broken link syntax)
  • Confirm tether export --help shows MODEL as positional and --output (not --out) in the real CLI
  • Confirm tether serve --help shows no --policy flag

🤖 Generated with Claude Code

…e what's-new, bad example flags)

- README.md line 16: "eleven verbs" -> "a focused set of verbs" (code has 18+
  visible top-level commands; the hardcoded count was wrong and brittle)
- README.md line 275: remove hardcoded "11 visible verbs" count for same reason
- README.md lines 14, 116, 484: de-link three dead hyperlinks to
  reflex_context/ (private vault not in this repo); keep surrounding prose,
  replace with inline backtick text noting the vault location
- README.md: add v0.12.0 "What's new" entry (Reflex->Tether rename,
  fastcrest-tether PyPI dist, reflex compat shim through v0.13.x, removed v0.14.0)
- CONTRIBUTING.md: replace nonexistent tether serve --policy smolvla --device cuda
  (--policy flag does not exist on serve; serve takes a positional export_dir) with
  tether go --model smolvla-base which is the correct one-command smoke test
- examples/02-deploy-smolvla-jetson.md line 65: fix tether export --model smolvla-base
  --out ./smolvla-export/ to tether export smolvla-base --output ./smolvla-export/
  (model is a positional arg, not --model; --out does not exist, correct flag is --output)
- THIRD_PARTY_LICENSES.md: update "Reflex VLA" -> "Tether" in project attribution
  header; update src/reflex/ -> src/tether/ path references for this project's own
  vendored files (upstream MIT license text and vendor name unchanged)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@rylinjames

Copy link
Copy Markdown
Collaborator Author

Reviewed the docs accuracy changes against the current CLI surface.

Verified locally:

  • smolvla-base resolves to lerobot/smolvla_base.
  • tether go --model is a valid command shape.
  • tether export MODEL --output is the correct export signature.
  • PYTHONPATH=src .venv/bin/python -m pytest tests/test_model_registry.py tests/test_one_command_deploy.py -p no:cacheprovider passes: 71 passed.

I cannot approve this PR through GitHub because it is authored by the same account, but the review is clean.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant