Audit a repository for AI-assisted-development risk without broad repository scanning or application-code changes.
Copilot ✓Cursor —Kilo —
AuthorNeexoCore
Updated
auditguardrails
Audit this repository for AI-assisted-development risk.
Do not scan the whole repository.
Inspect only:
root README or equivalent project overview
package metadata
top-level folder tree
.github configuration
existing AI instruction files
docs related to architecture, validation, deployment, or security
Identify:
missing or bloated AI instructions
generated, build, or asset folders that should be excluded
high-risk areas where agents should plan before editing
validation commands that should be documented
files that should be kept out of AI context
places where tenant, security, or database rules should be explicit
Do not edit code.
Return:
current state
recommended instruction files
recommended content exclusions
high-risk folders
manual GitHub admin settings
next PR plan
Raw content
Copy into your project — e.g. .instructions.md, .agent.md, or SKILL.md
Audit this repository for AI-assisted-development risk.
Do not scan the whole repository.
Inspect only:
- root README or equivalent project overview
- package metadata
- top-level folder tree
- `.github` configuration
- existing AI instruction files
- docs related to architecture, validation, deployment, or security
Identify:
- missing or bloated AI instructions
- generated, build, or asset folders that should be excluded
- high-risk areas where agents should plan before editing
- validation commands that should be documented
- files that should be kept out of AI context
- places where tenant, security, or database rules should be explicit
Do not edit code.
Return:
1. current state
2. recommended instruction files
3. recommended content exclusions
4. high-risk folders
5. manual GitHub admin settings
6. next PR plan