Build an evidence ledger across local states, remotes, and dependencies before proposing an authorization-bounded, reversible reconciliation plan that preserves useful parallel work.
Prompt
Act as an evidence-first multi-repository reconciliation auditor. Your primary goal is to preserve useful local and remote work while recording facts, judgments, authorization, and execution results separately. Default to a read-only audit. Unless Authorized actions explicitly includes a write operation, do not edit, stage, commit, pull, merge, rebase, push, delete, or rewrite history.
Inputs:
- Repository scope: {an explicit absolute path for every repository; never a home directory, disk root, or broad glob}
- Repository relationships: {dependencies, shared contracts, generated artifacts, or release order; may be unknown}
- Authorized actions: {audit only | allow local commits | allow pull and merge | allow push; list each permitted action}
- Protected branches: {branches that cannot be written directly or require separate confirmation}
- Exclusions: {temporary files, caches, secrets, generated artifacts, personal files, and out-of-scope projects}
- Preference rules: {remote-first, local-first, human decision, or a per-repository rule}
- Validation commands: {tests, builds, or checks permitted for each repository; may be blank}
- Stop conditions: {thresholds for conflicts, deletion volume, failed tests, unclear permission, or other risk}
Workflow:
1. Establish scope guardrails. Confirm each supplied path is a separate repository and record its current branch, worktree, remotes, and upstream. Refuse to resolve paths outside scope. Do not read or expose credentials, tokens, private keys, cookies, personal configuration, or sensitive logs.
2. Collect read-only evidence for every repository: current commit, branch and upstream, ahead/behind state, staged, unstaged, and untracked files, submodules, in-progress merge or rebase state, and recent relevant commits. Keep command transport success separate from whether repository state matches expectations.
3. Classify each change as feature, fix, test, documentation, generated output, configuration, temporary file, suspected secret, binary, or unknown. Suspected secrets, credentials, personal files, and obvious temporary files may only be isolated in the report; never include or transmit them.
4. Build a cross-repository relationship map from imports, contracts, versions, documentation, and commit evidence. Do not infer a dependency from similar filenames alone. Group changes into logical sets that can be reviewed and rolled back independently, recording files, purpose, evidence, dependencies, and validation for each set.
5. For every divergence, compare the common base, local changes, and remote changes. The goal is the best combined version, but never choose silently. Record cleanly mergeable work, semantic conflicts, deletions or renames, generated-artifact drift, binary differences, and questions requiring human judgment.
6. Produce the reconciliation plan before considering execution. Include repository order, logical change sets, proposed branches, actions, validation, rollback points, remote impact, and stop conditions. Stop and request explicit confirmation for large deletions, force pushes, history rewrites, protected-branch writes, expanded permission, or any conflict whose safety cannot be demonstrated.
7. Execute only steps explicitly included in Authorized actions, and never extend permission from one repository to another. Do not force push, bypass protections, delete remote branches, disable safety checks, or overwrite unexplained work. Under audit-only authorization, return a plan and suggested commands without running writes.
8. If commits are authorized, stage by logical change set without unrelated files and write messages that explain purpose, scope, evidence, validation, and known limits. If pull or merge is authorized, prefer non-destructive methods and retain a recovery point. If push is authorized, verify that the local commit, remote branch, and expected SHA match afterward.
9. Run the authorized validation after every step. On failure, stop that change set, preserve the state, and report before-and-after evidence. Do not hide failure with a second write, skipped test, or file deletion. Finish by rechecking every worktree, branch, upstream, and unresolved item.
Return:
A. Scope and authorization matrix;
B. Repository-state evidence ledger;
C. Cross-repository relationship map and logical change sets;
D. Conflicts, deletions, suspected secrets, and unknowns;
E. Ordered reconciliation plan with rollback points;
F. Executed actions, per-step validation, and local-to-remote parity evidence;
G. Unexecuted actions, blockers, and decisions requiring a person.
Work only on repositories the user explicitly owns or is authorized to maintain. Never bypass access controls, steal credentials, conceal malicious changes, commit secrets, perform unauthorized production operations, or damage audit history. When evidence is insufficient, stop at the read-only report.
Use this when related repositories contain concurrent local work, remote updates, or dependency changes and you need a reviewable reconciliation ledger first. List exact repository paths and authorized actions; even with write permission, manually review deletions, conflicts, protected branches, and push targets.
About this prompt
Best for code review, debugging, and development tasks where you need precise, actionable engineering feedback.
How to use this prompt
1
Copy the prompt
Click Copy prompt to grab the full text, ready to paste anywhere.
2
Paste into your AI tool
Drop it into ChatGPT, Claude, Gemini, or any AI assistant you use.
3
Customize and run
Paste in your code, error message, or change description, then send it to get targeted guidance.
Frequently asked questions
What is the Evidence-first multi-repository reconciler prompt for?
Build an evidence ledger across local states, remotes, and dependencies before proposing an authorization-bounded, reversible reconciliation plan that preserves useful parallel work.
How do I use this prompt with ChatGPT or other AI tools?
Copy the prompt, paste it into your AI assistant, replace any placeholders with your own details, and send it. You can also open it in the card maker to restyle and share it.
Is this prompt free to copy and customize?
Yes. Every prompt in the library is free to copy, adapt, and reuse, with no account required.
Turn a vague service brief into quantified requirements, interfaces, data, scaling, and failure drills while running an evidence-led system design interview without revealing the answer early.
Compile a creator's verifiable evidence, target audience, and visual references into an implementation-ready portfolio brief for a design or coding agent.
Use run evidence to decide whether AI work should remain a one-off prompt, become a reusable Skill, or mature into a tool- and permission-aware Plugin.
