Skip to content

install.sh member-set check rejects official v0.9.0 release assets (both standard and ui variants) #1510

Closed

Description

@viniciusdev772

Summary

install.sh on main fails on the official v0.9.0 release assets for both the --standard and --ui variants, with:

error: release archive does not match the exact standard member set

and

error: release archive does not match the exact ui member set

Checksum verification passes in both cases — the archive itself is intact and matches checksums.txt. The failure is install.sh's post-extraction member-count/name check, not a corrupted download.

Repro

curl -fsSL https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.sh | bash -s -- --standard
curl -fsSL https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.sh | bash -s -- --ui

Both fail after "Checksum verified." on Linux/amd64.

Root cause (as far as I can tell)

install.sh (main) expects:

  • standard: 5 members
  • ui: 6 members, including cbm-integrations.json and a cbm-ui-<sha256>.pack file

I downloaded and inspected the actual v0.9.0 assets directly (bypassing the installer):

codebase-memory-mcp-linux-amd64-portable.tar.gz (standard) contains only 4 members:

codebase-memory-mcp
LICENSE
install.sh
THIRD_PARTY_NOTICES.md

— missing cbm-integrations.json.

codebase-memory-mcp-ui-linux-amd64-portable.tar.gz (ui) also contains only 4 members — identical list, no cbm-integrations.json and no cbm-ui-*.pack at all. So the currently published ui asset doesn't actually differ from standard.

This looks like install.sh on main was updated for a newer release-asset layout (adding cbm-integrations.json and a content-addressed UI pack file) before a matching release was cut. The v0.9.0 assets (published 2026-07-08) still use the older 4-member layout.

Impact

The one-line installer in the README currently cannot install either variant on Linux/amd64 — it always exits 1 after checksum verification.

Suggested fix

Either:

  1. Cut a new release whose assets match the member set install.sh on main expects, or
  2. Pin install.sh's member-set check to what's actually in the latest published release until the new asset layout ships.

Happy to provide the full tar tvzf output for other platform archives if useful.

Activity

  1. added
    editor/integrationEditor compatibility and CLI integration
    parsing/qualityGraph extraction bugs, false positives, missing edges
    ux/behaviorDisplay bugs, docs, adoption UX
    on Aug 9, 2026
  2. archcorsair commented on Aug 9, 2026

    @archcorsair

    Confirming, I have the same issue.

    Linux 7.1.6-arch1-1 #1 SMP PREEMPT_DYNAMIC Tue, 04 Aug 2026 11:19:27 +0000 x86_64 GNU/Linux
    
  3. FoTrix commented on Aug 9, 2026

    @FoTrix

    Confirming this also affects macOS (Apple Silicon, arm64) with the --ui variant:

    codebase-memory-mcp installer
      os:      darwin
      arch:    arm64
      variant: ui
      target:  /Users/vios/.local/bin/codebase-memory-mcp
    
    Downloading codebase-memory-mcp-ui-darwin-arm64.tar.gz...
    Checksum verified.
    error: release archive does not match the exact ui member set
    

    I inspected the published v0.9.0 asset directly (codebase-memory-mcp-ui-darwin-arm64.tar.gz): it contains only 4 members — codebase-memory-mcp, LICENSE, install.sh, THIRD_PARTY_NOTICES.md — no cbm-integrations.json and no cbm-ui-*.pack. The v0.9.1-rc.1 asset has the same 4-member layout. So this is consistent with your analysis: install.sh on main enforces the new 5/6-member contract, but the last published releases still use the old 4-member layout.

    Workaround that worked for me on macOS: use the install.sh bundled inside the release archive (which predates the strict member check):

    curl -fsSL -o /tmp/cbm.tar.gz "https://github.com/DeusData/codebase-memory-mcp/releases/latest/download/codebase-memory-mcp-ui-darwin-arm64.tar.gz"
    mkdir -p /tmp/cbm && tar -xzf /tmp/cbm.tar.gz -C /tmp/cbm
    /tmp/cbm/install.sh --ui

    This installed codebase-memory-mcp 0.9.0 successfully and configured all detected agents.

  4. Vkozlove commented on Aug 10, 2026

    @Vkozlove

    I can confirm this issue on a second Apple Silicon Mac.

    The one-line installer from the README fails exactly as reported:

    curl -fsSL https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.sh | bash

    Result:

    Checksum verified.
    error: release archive does not match the exact standard member set
    

    The same happens with the --ui variant.

    However, I found that the release assets themselves appear to be fine. If I download the release archive and run the install.sh included inside that archive, installation completes successfully:

    curl -fsSL -o /tmp/cbm.tar.gz \
    "https://github.com/DeusData/codebase-memory-mcp/releases/latest/download/codebase-memory-mcp-ui-darwin-arm64.tar.gz"
    
    mkdir -p /tmp/cbm
    tar -xzf /tmp/cbm.tar.gz -C /tmp/cbm
    chmod +x /tmp/cbm/install.sh
    /tmp/cbm/install.sh --ui

    This installs correctly, configures the coding agents, and codebase-memory-mcp --version reports 0.9.0.

    That seems to confirm the issue is specifically a mismatch between the current main/install.sh and the published release assets, rather than a problem with the release archive itself.

    Until this is fixed, using the install.sh shipped inside the release archive seems to be a reliable workaround.

    Special thanks to user FoTrix

  5. added
    bugSomething isn't working
    priority/highNeeds near-term maintainer attention; high-impact bug, regression, safety issue, or release blocker.
    awaiting-reporterWaiting on the reporter for info/repro; stale bot will warn then close
    on Aug 10, 2026
  6. added this to the 0.10.0-rc milestone on Aug 10, 2026
  7. github-actions commented on Aug 10, 2026

    @github-actions

    Thanks for the report! To move this forward we need a bit more so we can reproduce it ourselves:

    • the codebase-memory-mcp --version you're on
    • the exact steps or command you ran
    • a public repo (or a small dummy snippet) that shows the problem — please don't paste proprietary code

    Once that's here we'll pick it straight back up. Heads-up: issues left awaiting-reporter are automatically closed after a few weeks of silence, but a comment reopens the door anytime.

  8. DeusData commented on Aug 10, 2026

    @DeusData
    Owner

    Thank you for the multi-platform reproductions and archive inspection. v0.10.0 is now published, and the current installer no longer contains the exact five-member or six-member checks reported here.

    Could the original reporter or one of the confirmers retry the same one-line installer with v0.10.0 on Linux or macOS and report the command and resulting version? I have marked this awaiting reporter and kept it in the immediate bug train until the released path is confirmed.

  9. archcorsair commented on Aug 10, 2026

    @archcorsair

    Thank you for the multi-platform reproductions and archive inspection. v0.10.0 is now published, and the current installer no longer contains the exact five-member or six-member checks reported here.

    Could the original reporter or one of the confirmers retry the same one-line installer with v0.10.0 on Linux or macOS and report the command and resulting version? I have marked this awaiting reporter and kept it in the immediate bug train until the released path is confirmed.

    Confirmed its working for me now, thank you

  10. DeusData commented on Aug 12, 2026

    @DeusData
    Owner

    Fixed on main by @cyphercodes' #1514 (merge 60a2bbf) — shipping in v0.10.3.

  11. FoTrix commented on Aug 12, 2026

    @FoTrix

    everything, correct and arranged. Thank you so much!

  12. DeusData commented on Aug 12, 2026

    @DeusData
    Owner

    Thank you for confirming the repaired installer path, @FoTrix. That closes the loop cleanly; this issue does not need to be reopened.

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

Metadata

Metadata

Assignees

No one assigned

    Labels

    awaiting-reporterWaiting on the reporter for info/repro; stale bot will warn then closebugSomething isn't workingeditor/integrationEditor compatibility and CLI integrationparsing/qualityGraph extraction bugs, false positives, missing edgespriority/highNeeds near-term maintainer attention; high-impact bug, regression, safety issue, or release blocker.ux/behaviorDisplay bugs, docs, adoption UX

    Projects

    No projects

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions