Skip to content

security: publish the Symfony H3 decision packet - #17

Open
phibkro wants to merge 1 commit into
publish/0014-tutor-tracerfrom
publish/0015-h3-packet
Open

security: publish the Symfony H3 decision packet#17
phibkro wants to merge 1 commit into
publish/0014-tutor-tracerfrom
publish/0015-h3-packet

Conversation

@phibkro

@phibkro phibkro commented Aug 12, 2026

Copy link
Copy Markdown
Member

Design spec

design-specs/0015-security-h3-decision-packet.md

Journey

A maintainer generates a deterministic H3 security decision packet from the current Symfony security surface, with route closure evidence, unresolved risks, and an explicit decision boundary.

Experience it

  1. Check out this stacked branch.
  2. Run the packet/report/check commands recorded in the spec.
  3. Inspect the generated decision packet and route-security inventory.

Evidence

  • Focused report/build/typecheck/lint and deterministic digest gates: PASS.
  • Independent code, security, and runtime review: PASS.
  • Exact 16-path capsule including source, tests, report, and final spec.

What is real

The security inventory and decision packet are reproducible. The substantive H3 production decision is still unresolved; this PR changes no runtime security policy, provider, credentials, deployment, or production behavior.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a19d12f127

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment on lines +1252 to +1253
const manifest = fixtureManifestCanonical();
if (await sha256Prefixed(manifest) !== FIXTURE_MANIFEST_SHA256) {

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Read and hash the supplied fixture manifest

When --fixture-manifest-path names a missing or tampered file inside the allowed capsule directory, this branch ignores the file and hashes the generator's hard-coded fixtureManifestCanonical() value instead. The command therefore exits successfully and publishes a passing receipt even for a nonexistent manifest, so H3_FIXTURE_MANIFEST_DRIFT cannot validate the artifact the receipt claims to use.

Useful? React with 👍 / 👎.

Comment on lines +1203 to +1208
const localeInput = ["ä", "a", "z", "Å"];
const localeA = [...localeInput].sort(new Intl.Collator("sv").compare);
const localeB = [...localeInput].sort(new Intl.Collator("en").compare);
const canonicalA = canonicalJson({ values: [...localeA].sort(compareCodeUnits) });
const canonicalB = canonicalJson({ values: [...localeB].sort(compareCodeUnits) });
const f13 = result("F13_locale_order", "H3_NONDETERMINISTIC_OUTPUT", localeA.join("|") !== localeB.join("|") && canonicalA === canonicalB && await sha256(canonicalA) === await sha256(canonicalB));

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Exercise packet generation in the locale falsifier

F13 can pass even if decision-packet generation becomes locale-dependent because it compares only a hard-coded four-string array and never builds or hashes a packet under either locale. Consequently the emitted frozen falsifier receipt asserts H3_NONDETERMINISTIC_OUTPUT was tested without exercising the output whose determinism matters.

Useful? React with 👍 / 👎.

Comment on lines +1063 to +1065
const required: Record<string, ReasonCode> = { F1_missing_route: "H3_LEGACY_CANDIDATE_MISSING_ROUTE", F2_new_current_operation: "H3_CURRENT_OPERATION_UNSEEN_IN_POLICY", F3_method_change: "H3_METHOD_MISMATCH", F4_duplicate_owner: "H3_DUPLICATE_OPERATION", F5_unknown_method: "H3_METHOD_UNRESOLVED", F6_get_mutates: "H3_GET_SIDE_EFFECT", F8_count_drift: "H3_POLICY_COUNT_MISMATCH", F9_identity_leak: "H3_PII_INPUT", F14_resource_key_wrong_kind: "H3_KEY_KIND_MISMATCH", F15_resource_key_method: "H3_KEY_KIND_MISMATCH" };
const reason = required[caseId];
const pass = packetReasons.includes(reason) && packetResult.packet.recommendation === "fail_closed" && (packetResult.packet.status === "invalid" || caseId === "F1_missing_route" || caseId === "F2_new_current_operation" || caseId === "F3_method_change" || caseId === "F4_duplicate_owner" || caseId === "F5_unknown_method" || caseId === "F6_get_mutates");

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Scope the duplicate-operation falsifier to its target

F4 checks only whether the packet-wide reason catalog contains H3_DUPLICATE_OPERATION. The unmutated synthetic inventory already produces that reason because operations 224–225 share all_departments and 226–228 share survey_admin, so F4 still reports pass if the intended duplicate of fixture:route:1 is not detected. Verify the target row and both expected operation IDs rather than the aggregate reason list.

Useful? React with 👍 / 👎.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant