anthropics / anthropics/claude-code

[Feature Request] Add developer context awareness for local debugging scenarios

Offen
#93,346 0 Kommentare 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen
area:model enhancement platform:macos
Vorherrschende Sprache
Python
Sterne
145k
Forks
23.1k
PR-Merge-Kennzahlen
PR-Kennzahlen ausstehend

Beschreibung

**Bug Description**
Dear Anthropic Support / Moderation Team,

I am writing to appeal a recent restriction on my session/account regarding the use of Claude Opus.

While developing and debugging a web application locally, I accidentally locked myself out of my administrative account because an API endpoint (/update-password) unexpectedly updated the admin password during local front-end testing.

To resolve this developer-side roadblock, I asked Claude to suggest potential default values or assist in resetting/generating a replacement password string for my development database. Unfortunately, the prompt wording was flagged by the automated safety system for potential security policy violations (e.g., password guessing/unauthorized access).

I want to clarify that:

This was purely a legitimate developer debugging task in my own controlled environment.

There was no intent to perform unauthorized access, credential harvesting, or malicious hacking.

It was an automated false positive caused by the context of my request.

I fully understand and respect Anthropic's safety policies, and I will be more careful with how I phrase code/database management requests in the future. Could you please review this prompt and restore my access to Claude Opus?

Thank you for your time and assistance.

Best regards,

**Environment Info**
- Platform: darwin
- Terminal: Apple_Terminal
- Version: 2.1.263
- Feedback ID: 811856bb-59f8-4b2c-82e3-fe6d3e7f2d7a

**Errors**
```json
[]
```

Beitragsleitfaden

Für dieses Repository ist kein Beitragsleitfaden indexiert

Rechercherichtung

The report names no repository files, tests, or entry points to inspect. It describes an account restriction and a moderation false positive rather than a defined code change, so the intended behavior and completion criteria need clarification before implementation.

Vom Indexierungsmodell aus dem Issue-Text verfasst.

Bewertung

Bereich
security
Issue-Typ
Feature
Schwierigkeit
5/5
Geschätzter Aufwand
Über eine Woche
Aktivitätsstatus
Aktiv
Klarheit
Muss geklärt werden
Anfängerfreundlichkeit
20/100

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.