Fallback procedures when the Hermes ↔ Codex memory bridge writes .Codex/memory outputs but the internal git commit/push path fails because of dirty, stale, or broken submodule state.
Scanned 9/9/2026
Install to Claude Code
npx -y skills add vamseeachanta/workspace-hub --skill memory-bridge-commit-fallbacks --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Memory Bridge Commit Fallbacks?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/vamseeachanta-memory-bridge-commit-fallbacks)More formats (shields.io, HTML) on the badges page.
---
name: memory-bridge-commit-fallbacks
description: Fallback procedures when the Hermes ↔ Codex memory bridge writes .Codex/memory outputs but the internal git commit/push path fails because of dirty, stale, or broken submodule state.
---
# Memory Bridge Commit Fallbacks
## When to use
Use when running `scripts/memory/pre-bridge-quality.sh --fix` or `scripts/memory/bridge-hermes-Codex.sh --commit` in workspace-hub and the bridge successfully updates `.Codex/memory/` files, but its internal commit/push phase fails due to unrelated dirty repo state, stale/broken submodule status, or errors such as:
- `fatal: 'git status --porcelain=2' failed in submodule ...`
- `error: status died of signal 15`
- bridge output shows files updated, then fails while stashing/pulling/committing
This is a fallback for landing the bridge outputs only; do not use it to commit unrelated repo churn.
## Procedure
1. Confirm the quality gate passed and was not a degenerate-memory abort.
- If score `< 50`, do **not** bridge.
- If score `>= 50`, continue only if `.Codex/memory/` outputs were written.
2. Verify remote relationship before creating a manual commit:
```bash
cd /mnt/local-analysis/workspace-hub
git fetch origin main
git rev-list --left-right --count origin/main...HEAD
```
Expected safe case: `0 0` or only local bridge work pending. If behind, avoid broad rebase/stash in a dirty checkout; resolve upstream state first or use a clean worktree.
3. Inspect only memory outputs, not repo-wide churn:
```bash
git diff --cached --name-only -- .Codex/memory
git status --short --ignore-submodules=all -- .Codex/memory
```
4. Commit with an explicit pathspec so unrelated staged/dirty files are excluded:
```bash
git commit -m "chore(memory): auto-refresh memory bridge ($(date +%F))" -- .Codex/memory
```
5. Push and verify remote HEAD:
```bash
git push origin main
git rev-parse HEAD
git ls-remote origin refs/heads/main | cut -f1
```
6. Verify the bridge succeeded semantically:
```bash
bash scripts/memory/check-memory-drift.sh
```
Expected result: `In sync — no drift detected` and exit code `0`.
7. Report honestly:
- quality score
- drift count from the initial drift check
- `.Codex/memory/` files updated with line counts
- that the internal script commit failed and the path-limited fallback commit/push was used
- final commit SHA and drift recheck result
## Pitfalls
- Do not run broad `git stash`, `git add .`, or repo-wide `git commit` in workspace-hub when unrelated agent/session churn is present.
- Do not trust repo-wide `git status` if submodules are broken; use `--ignore-submodules=all` and path-limited checks for `.Codex/memory/`.
- Do not rerun the bridge repeatedly after outputs were written; repeated runs can change timestamps and expand the diff.
- This skill overlaps with `memory-bridge-operation`; prefer updating that canonical skill when it is editable.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!