Skip to content

data-safeguards.md

Latest commit

 

History

History
88 lines (60 loc) · 3.08 KB

data-safeguards.md

File metadata and controls

88 lines (60 loc) · 3.08 KB
Raw

Security & Data Safeguards (Developer Policy)

This file is the developer-level implementation policy that aligns with docs/security.html.

Non-negotiable baseline

No private, protected, or production data may be committed to this repository.

This includes direct values, screenshots, logs, exports, copied snippets, and any derived artifact that can identify real people, systems, or internal infrastructure.

What is allowed vs prohibited

Category Allowed Prohibited
Screenshots Anonymized, redacted captures Unredacted UI with names, emails, IDs, case data
Data Synthetic/dummy records Production or user-originated records
Schemas Field names/types/relations Real data values embedded in examples
Infrastructure Generic architecture patterns Internal hostnames, IP ranges, private endpoints
Secrets Placeholders and mock values API keys, passwords, tokens, live certs

Secure workflow by phase

Phase 1 (Discovery)

  • Redact screenshots before adding to LotusApp/screenshots/
  • Keep only structural schema details in LotusApp/data-schema/
  • Do not paste sensitive details into Copilot prompts

Phase 2 (Requirements)

  • Write requirements in business-functional language
  • Exclude references to internal production systems and addresses
  • Keep real case/user examples out of user stories

Phase 3 (Planning)

  • Use generic environment naming (e.g., dev, test, prod)
  • Do not include tenant identifiers, secret names, or private URLs in architecture docs
  • Document security assumptions explicitly

Phase 4 (Execution)

  • Keep secrets in environment variables or secure secret stores only
  • Never hardcode credentials in code, tests, config, or markdown
  • Prefer fake fixtures and synthetic test datasets

Redaction standard

Before committing an image or copied snippet, verify:

  • Names, emails, usernames, IDs, and case numbers are removed
  • Dates/timestamps that can identify records are generalized if sensitive
  • Internal URL bars, hostnames, and environment labels are hidden
  • Any browser tab/title that reveals internal systems is masked

If unsure, do not commit until reviewed by a teammate.

Commit-time checklist (required)

  • No real user or production data in staged files
  • No credentials or tokens in staged files
  • No internal endpoints/IPs/hostnames in staged files
  • Screenshots are redacted
  • Dummy/test data is clearly synthetic
  • AI-generated code was reviewed by a human before commit

If sensitive content is discovered

  1. Stop sharing and stop further commits that propagate exposure.
  2. Notify project owners/security contacts immediately.
  3. Rotate any potentially exposed credentials immediately.
  4. Remove exposure from history using approved org process/tooling.
  5. Document what happened and the corrective action taken.

Do not treat file deletion alone as remediation; history must be addressed.

Developer operating principle

Use AI to accelerate implementation, not to relax controls.
When in doubt: leave it out and ask before committing.