Security Grades
Samchil grades are not a marketing layer. Moving up a tier never means “more features” — it means more defense. A grade is a ✓ matrix showing how far a module goes across four universal axes. No score math: you read, at a glance, what it blocks.
Naming reflects security state — public verification → hardening → audited assurance — not a feature tier. Each grade contains every defense of the one below it, and only stacks more security on top.
FREE
We publish how it works. Read the source and verify it yourself — proof of direction, not a demo.
HARDENED
Blocks the attacks that get exploited most in the real world — across all 4 axes. Ship it straight to production.
AUDITED
Beyond blocking — proof of what happened. Dual verification, audit report, and CVE response policy included.
Not a specific technique (HMAC, nonce…) but defense areas common to every backend security module. Each axis is anchored where real breach statistics concentrate.
Axis 1
Does it verify who the caller is and what they are allowed to do?
NoneNo verification.
BasicCaller identity verified — keys, tokens, or signatures, whichever fits the module.
FullIdentity + scope and object-level permissions, least-privilege enforced.
Axis 2
Does it control the data and the volume of calls coming in?
NoneNo control.
BasicSchema / type validation across the whole structure.
FullSchema + rate & resource limits + whitelisting to block abuse patterns.
Axis 3
Do errors and retries avoid leaking information or corrupting state?
NoneRaw errors exposed; retries duplicate or corrupt.
BasicError masking + idempotency — duplicate requests stay safe.
FullIdempotency + audit trail + per-environment error isolation (dev/prod auto-switch).
Axis 4
Can you trace what happened, and is there a structure to take responsibility?
NoneNo logs, no response policy.
BasicCore event logging + request-freshness checks (expiry, timestamps).
FullAudit-trail logging + single-use guarantee (nonce) + public CVE response policy.
FREE proves the direction with the core algorithm only. HARDENED clears Basic on all four axes — production-ready. AUDITED clears Full on all four — auditable trust.
| Judgment axis | FREE | HARDENED | AUDITED |
|---|---|---|---|
| 1 · Access Control | — | ✓ Basic — identity verification | ✓ Full — scope & object permissions |
| 2 · Input & Abuse Defense | — | ✓ Basic — schema validation | ✓ Full — + rate limit & whitelist |
| 3 · Safe Failure | — | ✓ Basic — idempotency, masking, safe retry | ✓ Full — audit trail & env isolation |
| 4 · Accountability | — | ✓ Basic — core logging & freshness | ✓ Full — audit trail, nonce, CVE policy |
To earn a grade, a module must satisfy every axis cell in that grade’s row. Miss a single axis and the module drops to the grade below.
This mirrors the first principle of security — one breach anywhere makes the whole thing meaningless. The old weighted-sum scoring is gone; the rule is built into the matrix itself.
Axes outside a product’s responsibility are marked N/A (with a one-line rationale in the design doc) and excluded from judgment — never used to dodge a hard requirement.