- Home
- Skills
- Testing & Debugging
- CtrlZ Repository Recovery Audit
Works with the AI Agents you already use
CtrlZ Repository Recovery Audit
Diagnose broken builds and unfinished repositories with source evidence, dependency checks, reproducible baselines, and before-and-after repair verification.
$10
CtrlZ Repository Recovery Audit
Example session with this skill installed
Inspect this repository and diagnose why the queue service retries after reaching its retry budget. Find the correct test command and working directory, reproduce the failure, apply a minimal fix, and verify it against the original tests. Report the evidence and any remaining coverage gaps.
- Read your context and instructions
- Compiled the ctrlz repository recovery
Recovery report
Verdict — Verified within the tested scope
This example uses a synthetic Python queue-service fixture.
Diagnosis
The Makefile identified service/ as the working directory and Python unittest as the test runner.
The baseline collected two cases. test_retry_under_budget passed, while test_retry_at_budget failed because the retry predicate returned True when the current attempt equalled the maximum.
Source inspection located the boundary error in service/queue_policy.py:3.
Repair
The predicate used an inclusive comparison:
return attempt <= max_attempts
The repair changed it to
return attempt < max_attempts
Verification
Executed from service/:
python3 -m unittest -v test_queue_policy
| Check | Before | After |
|---|---|---|
| Exit code | 1 | 0 |
| Passed | 1 | 2 |
| Failed | 1 | 0 |
| Skipped | 0 | 0 |
The original test file checksum and collected test identities matched across both runs.
Coverage and next steps
Verification covered the retry predicate’s boundary behavior. Queue transport, scheduling, and production behavior were outside this fixture.
No dependency advisory lookup was performed. The fixture used Python’s standard library.
Review the patch and repeat the checks in the intended environment before release.
Connects securely to your tools. The creator never sees your data.
About this skill
CtrlZ Repository Recovery Audit guides your coding agent through investigating a broken or unfinished repository.
It identifies the relevant package, runtime, setup, and test commands from repository evidence. It reviews source findings, checks resolved dependencies when advisory access is available, and establishes a test baseline when execution is permitted.
When you request a repair, it guides the agent through a focused change and repeats the original checks to compare results, test identities, and skipped cases.
You receive a structured recovery report with findings, reproduction commands, proposed next actions, verification results, and remaining coverage gaps.
Requires a coding agent with access to the repository. Test execution depends on the available runtime and execution environment. Dependency advisory checks require network access. Findings are bounded by the files and checks actually reviewed.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Testing & Debugging.
Results depend on the coding agent, repository access and available execution environment. Inspection guidance covers Python and Node.js projects most directly. Runtime tests may be blocked by missing dependencies, unsupported tooling or unavailable isolation. Advisory coverage depends on resolved package versions and provider availability. Passing checks establish only the behavior tested. Comprehensive security certification and guaranteed repairs are outside the scope.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 11 days ago
- Software
- Terminal accessProject runtimeIsolated execution environment
Python
The project runtime runs the tests. Run unfamiliar code only in a permitted isolated environment.
Permissions
File access is limited to the repository selected by the user. Inspection reads source, manifests, lockfiles, tests and CI configuration. Writing supports reports and user-requested repairs. Terminal access supports permitted baseline and verification checks. Optional advisory queries send package names, ecosystems and resolved versions. Credential files, dependency caches and paths outside the selected repository are excluded from inspection.