Skip to content

docs(readme): present product as OpenAB Computer during rename - #64

Merged
chaodu-agent merged 1 commit into
mainfrom
docs/openab-computer-rebrand-readme
Oct 29, 2026
Merged

chaodu-agent merged 1 commit into
mainfrom
docs/openab-computer-rebrand-readme

Conversation

@chaodu-agent

Copy link
Copy Markdown
Contributor

What problem does this solve?

The product is being renamed from instance-mcp to OpenAB Computer, tracked in #63. The README still presents the product only as oab-instance-mcp. This PR updates the title and one prose mention, and adds a note that the rename is in progress.

Refs #63

Discord Discussion URL: Not applicable - maintainer-filed docs step for the rename tracked in #63.

Review Contract

Goal

Make the README present the product as OpenAB Computer while the executable, paths, LaunchAgent label, and bundle id still use their current names.

Non-goals

  • No change to commands, install paths, config paths, log paths, bundle id, LaunchAgent label, Keychain or TCC identifiers, or release asset names.
  • No change to the repository name or URLs. The repository is renamed in a later step.
  • No change to code, scripts, or other docs.

Accepted Residual Risks

  • The title and the commands now use different names until the rename ships. The note at the top of the README says so. This is intended and temporary.

Acceptance Criteria

Follow-ups

At a Glance

README title:   oab-instance-mcp  ->  OpenAB Computer (oab-instance-mcp)
README prose:   instance-mcp      ->  OpenAB Computer
Commands/paths: unchanged

Prior Art & Industry Research

Not applicable - documentation-only change with no behavior change.

Title and the TTL prose now name OpenAB Computer. Commands, paths,
bundle id, and release assets keep their current names until the
rename ships. Refs #63
@chaodu-agent
chaodu-agent merged commit 5360597 into main Oct 29, 2026
7 checks passed
@chaodu-agent
chaodu-agent deleted the docs/openab-computer-rebrand-readme branch October 29, 2026 16:32
@chaodu-agent

Copy link
Copy Markdown
Contributor Author

Important

CHANGES REQUESTED ⚠️ (post-merge review, informational only, blocks nothing) - The title and TTL prose correctly switch to OpenAB Computer, but the new rename banner links the old repository URL and implies the executable and bundle id will be renamed later, and five older repository URLs remain.

Label: post-merge review. PR 64 is already merged into main as squash commit 53605979ee6e9a90dd6983f492924634305be001. This verdict is informational. It does not block anything and no commit status applies. The findings below are suggestions for a follow-up PR.

What This PR Does

The README now presents the product as OpenAB Computer while the repository moves from openabdev/instance-mcp to openabdev/openab-computer (tracked in #63). Operational identifiers (executable oab-instance-mcp, bundle id / LaunchAgent label dev.openab.instance-mcp, config paths, release asset names) are deliberately left unchanged.

How It Works

A prose-only change to README.md (+8/-2):

  • The H1 becomes OpenAB Computer (oab-instance-mcp), pairing the product name with the unchanged executable name.
  • A "Renaming in progress" banner is added at lines 3-7.
  • The TTL sentence at line 201 now says "OpenAB Computer accepts up to 24h" instead of "instance-mcp accepts up to 24h".

No command, path, identifier or asset name in the README was changed.

Findings

# Severity Finding Location
F1 🟡 The issue link in the newly added banner points at the old repository URL README.md:5
F2 🟡 Banner wording ("Until that rename ships ... keep their ... names") implies the executable, paths, label and bundle id will be renamed later; they are meant to stay unchanged, and the repository rename has already happened README.md:3-7
F3 🟡 Five pre-existing repository URLs still use openabdev/instance-mcp (outside this PR's stated scope) README.md:77, :90, :92, :187, :271
F4 🟢 Title and TTL prose renamed correctly; no command, path, bundle id or asset name was wrongly renamed README.md:1, :201
F5 🟢 Small, focused, single-file change with an accurate commit message -
Finding Details

🟡 F1: New banner link uses the old repository URL

Evidence (re-verified at the merge commit): git diff P M shows line 5 was added by this PR: > [#63](https://github.com/openabdev/instance-mcp/issues/63)). Until that rename ships, the. Line 4, one line above, names the new repository openabdev/openab-computer. The local clone's origin is already https://github.com/openabdev/openab-computer.git, so the rename has happened and this link works only through GitHub's redirect.
Impact: A line introduced by the rename PR itself uses the URL the rename is meant to replace. If a new repository named instance-mcp is ever created under the org, the redirect breaks.
Suggested fix: [#63](https://github.com/openabdev/openab-computer/issues/63).

🟡 F2: Banner implies identifiers will be renamed later

Evidence (re-verified at the merge commit): Lines 3-7 read: "Renaming in progress. ... Until that rename ships, the executable, install paths, LaunchAgent label and bundle id keep their oab-instance-mcp / dev.openab.instance-mcp names." The squash commit message repeats this ("keep their current names until the rename ships"). The rename decision for this work is that the executable, bundle id, config paths and release asset names do not change; only repository URLs and product-name prose change.
Impact: Readers may expect oab-instance-mcp, ~/.config/oab-instance-mcp/ and the TCC bundle id to change in a future release, and may delay scripts or TCC grants or plan a migration that will not happen. "Until that rename ships" is also stale because the repository rename has shipped.
Suggested fix: Reword along these lines: "Renamed. This product is now OpenAB Computer (repository openabdev/openab-computer, see #63). The executable oab-instance-mcp, its install and config paths, the LaunchAgent label / bundle id dev.openab.instance-mcp, and release asset names are unchanged and will stay that way."

🟡 F3: Remaining old repository URLs

Evidence (re-verified at the merge commit): git show M:README.md | grep -n openabdev/instance-mcp returns, besides line 5:

  • :77 [Releases](https://github.com/openabdev/instance-mcp/releases)
  • :90 https://api.github.com/repos/openabdev/instance-mcp/releases/latest
  • :92 https://github.com/openabdev/instance-mcp/releases/download/v$V/oab-instance-mcp-$V-linux-$A.tar.gz
  • :187 [#45](https://github.com/openabdev/instance-mcp/issues/45)
  • :271 [GitHub Releases](https://github.com/openabdev/instance-mcp/releases)

These lines predate the PR, which limited itself to the title and TTL sentence, so this is scope carry-over rather than a defect in the diff. They still work today through GitHub's redirect.
Downgraded sub-claim: One reviewer stated that the line 92 download (curl -fsSLO) does not follow redirects. That does not reproduce: -fsSLO is the combined flags -f -s -S -L -O, so -L is present, as it is on line 90 (-fsSL). The claim is dropped; the finding stays at 🟡 for consistency, not for a broken install command.
Suggested fix: In a follow-up PR, change only the repository segment to openabdev/openab-computer (and api.github.com/repos/openabdev/openab-computer). Keep the asset filename oab-instance-mcp-$V-linux-$A.tar.gz on line 92 and cd oab-instance-mcp-$V-linux-$A on line 93 exactly as they are.

🟢 F4: Product name changed correctly, identifiers untouched

The H1 (:1) and TTL sentence (:201) change only product-name prose. The remaining oab-instance-mcp / dev.openab.instance-mcp occurrences are executable, path, bundle-id or asset-name uses that must stay, and none was renamed. The only bare instance-mcp left in prose is line 3, where it names the old product on purpose.

🟢 F5: Scope and commit message

One file, +8/-2. The squash message accurately describes what changed and references #63.

Reviewer Summary
Reviewer Status Verdict Findings
Reviewer A ok CHANGES REQUESTED (informational) F1 + F3 (reported as one URL finding); redirect sub-claim dropped as not reproducible
Reviewer B ok CHANGES REQUESTED (informational) F1, F2, F3, F4, F5

This round used two reviewers. All 🟡 findings were re-verified against the merge commit with git show / git diff / grep.

Baseline Check
  • Review type: post-merge review, read-only. No commit status applies.
  • PR: docs(readme): present product as OpenAB Computer during rename #64 (repository renamed from openabdev/instance-mcp; GitHub redirects)
  • PR opened: 2026-10-29T16:08:29Z; merged: 2026-10-29T16:32:12Z
  • Merge commit (M): 53605979ee6e9a90dd6983f492924634305be001 (squash, single parent)
  • Parent (P = M^1): de5e211378692344b02781273a8437808fc62f8a
  • PR head sha: 867663e335972bc8000e6366d1b20dd3ea4f8628 (not present in the local clone; the review used git diff P M, which matches the PR file list)
  • Base branch: main
  • Diff stat (git diff --stat P M): README.md | 10 ++++++++--, 1 file changed, 8 insertions(+), 2 deletions(-)
  • Main already had: all commands, the executable and asset names, bundle id, config paths, and the old repository URLs.
  • Net-new value: the OpenAB Computer title, the rename banner, and the product-name fix in the TTL sentence.
What's Good (🟢)
  • No command, path or identifier was wrongly renamed, so copy-pasted README commands still work.
  • Keeping the executable name in the H1 helps people who search for the binary name.
  • The scope is minimal and easy to verify.

Devil's Advocate: Three Reasons We Might Not Need This PR

  1. Partial rename creates mixed naming - It changes two prose spots, leaves five old repository URLs in place and adds a sixth (F1), so readers see both names until a follow-up lands. One full-README pass would have avoided the intermediate state.
  2. Redirects already cover the URLs - GitHub redirects the old repository, and the executable name already identifies the project, so the practical gain of a prose-only rename is small.
  3. The banner adds churn - The "Renaming in progress" banner is temporary wording that already needs another edit (F2), and the TTL edit is cosmetic and changes no behavior.

These points do not argue for reverting the merge. They argue for one follow-up PR that fixes F1-F3 together.

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