design security scan and fixes, basically all trusted vs tainted input

2026-05-25 01:28:43 -06:00
parent 44d9d2065c
commit d15e09504a
5 changed files with 56 additions and 22 deletions
@@ -17,8 +17,10 @@
 ## Edge Cases

 - Bundle does not parse -> hard stop the slice because no trustworthy ingest artifacts can be emitted.
+- Bundle exceeds configured size or parse budget -> hard stop the slice before deeper ingest work to reduce resource-exhaustion risk.
 - An apparent boundary cannot be proven deterministically -> do not emit it as a separate segment record; keep only proven AST slice boundaries.
 - Identical upstream snapshot is ingested again -> derive the same run identity inputs and emit the same deterministic machine artifacts.
+- Previous run manifest is available but not verified -> do not use it for continuity hints.
 - Previous run manifest is available but continuity is weak -> ingest may observe it for continuity hints, but the current run artifacts still come only from the current snapshot and deterministic ingest rules.
 - Human-readable summary generation fails -> slice still succeeds if machine-readable artifacts were emitted correctly.

@@ -28,6 +30,8 @@
 - Rule: A `Segment Record` may come from any deterministic AST slice boundary that can be proven stably.
 - Rule: Ingest emits machine-readable artifacts as the source of truth for later phases.
 - Rule: Human-readable summary output is optional relative to core ingest success.
+- Rule: Bundle location input must be parsed into a trusted bundle location before ingest uses it.
+- Rule: A previous run manifest must be verified before it may influence continuity hints.
 - Invariant: If the bundle cannot be parsed, ingest hard-stops rather than emitting speculative artifacts.
 - Invariant: The workflow does not guess segment boundaries that it cannot prove deterministically.
 - Invariant: Identical upstream snapshot inputs produce identical deterministic ingest outputs.
@@ -14,7 +14,7 @@ State owned by `ingest-snapshot` and required to decide this workflow:

 - `SelectedSnapshot`
  - upstream snapshot identity
-  - upstream bundle location or source reference
+  - trusted bundle location derived from tainted ingest input
  - optional upstream metadata intended for later release provenance
 - `RunIdentityRules`
  - deterministic derivation rules from upstream snapshot identity
@@ -30,7 +30,7 @@ State owned by `ingest-snapshot` and required to decide this workflow:

 Snapshots or handoffs read but not owned by this context:

- optional previous `Run Manifest` reference used only for continuity hints
+- optional verified previous `Run Manifest` reference used only for continuity hints
 - upstream metadata needed later by `release-packaging`

 ## Policy Signature (Pseudo)
@@ -38,6 +38,12 @@ Snapshots or handoffs read but not owned by this context:
 ```text
 deriveRunIdentity : SelectedSnapshot -> RunIdentity

+parseBundleLocation :
+  TaintedBundleInput -> Result<TrustedBundleLocation, IngestRejected>
+
+validatePreviousRunManifest :
+  RunManifest -> Result<VerifiedPreviousRunManifest, IngestRejected>
+
 validateSnapshotIngest :
  IngestUpstreamSnapshot -> SelectedSnapshot -> Result<SnapshotReady, IngestRejected>

@@ -77,5 +83,6 @@ performSnapshotIngest :
 - The `ingest-snapshot` context decides only whether one snapshot can be deterministically ingested and which boundaries become `Segment Records`.
 - It does not decide package identity, dependency externalization, context heuristics, lineage matching, naming, regularization, transform replay, or release publication.
 - Observing a previous `Run Manifest` does not let this slice reuse or override current ingest decisions; cross-run lineage belongs to `snapshot-lineage`.
+- A previous `Run Manifest` must be verified for schema and integrity before this slice may use it as a continuity hint.
 - Human-readable summary generation is outside the hard success contract for this slice; required machine artifacts remain the source of truth.
 - Feature-level orchestration decides when later phases may continue after downstream review-needed states; this slice only hard-stops when deterministic ingest itself is not trustworthy.
@@ -4,10 +4,14 @@ open IngestSnapshot.SharedModel

 // 1. Primitives

-type TaintedBundleInput = TaintedBundleInput of BundleSource
+type TaintedBundleInput = TaintedBundleInput of TaintedBundleLocation

 type DerivedRunIdentity = DerivedRunIdentity of RunIdentity

+type MaxBundleBytes = MaxBundleBytes of int64
+
+type ParseBudget = ParseBudget of int64
+
 type BoundaryProof = BoundaryProof of string

 type RequiredArtifact =
@@ -18,6 +22,9 @@ type RequiredArtifact =
 type IngestFailureReason =
    | BundleNotParseable
    | RunIdentityCouldNotBeDerived
+    | PreviousRunManifestNotVerified
+    | BundleTooLarge of MaxBundleBytes
+    | ParseBudgetExceeded of ParseBudget
    | NoDeterministicBoundaryProven
    | RequiredArtifactMissing of RequiredArtifact

@@ -27,15 +34,15 @@ type IngestUpstreamSnapshot =
    { SnapshotIdentity: SnapshotIdentity
      BundleInput: TaintedBundleInput
      SnapshotMetadata: SnapshotMetadata option
-      PreviousRunManifest: RunManifest option }
+      PreviousRunManifest: VerifiedPreviousRunManifest option }

 // 3. Events (Facts)

 type UpstreamSnapshotIngested =
    { RunManifest: RunManifest
      SegmentRecords: SegmentRecord list
-      CanonicalProjectionPath: CanonicalProjectionPath
-      SummaryPath: SummaryPath option }
+      CanonicalProjectionPath: TrustedCanonicalProjectionPath
+      SummaryPath: TrustedSummaryPath option }

 type SnapshotIngestHardStopped =
    { SnapshotIdentity: SnapshotIdentity
@@ -54,17 +61,21 @@ type Error =
 type AwaitingSnapshotSelection =
    { RunIdentityRulesDescription: string
      BoundaryRulesDescription: string
-      RequiredArtifacts: RequiredArtifact list }
+      RequiredArtifacts: RequiredArtifact list
+      MaxBundleBytes: MaxBundleBytes
+      ParseBudget: ParseBudget }

 type SnapshotReady =
    { SelectedSnapshot: SelectedSnapshot
-      PreviousRunManifest: RunManifest option
-      RequiredArtifacts: RequiredArtifact list }
+      PreviousRunManifest: VerifiedPreviousRunManifest option
+      RequiredArtifacts: RequiredArtifact list
+      MaxBundleBytes: MaxBundleBytes
+      ParseBudget: ParseBudget }

 type DeterministicSegmentsReady =
    { RunIdentity: RunIdentity
      SelectedSnapshot: SelectedSnapshot
-      PreviousRunManifest: RunManifest option
+      PreviousRunManifest: VerifiedPreviousRunManifest option
      SegmentRecords: SegmentRecord list
      BoundaryProofs: BoundaryProof list
      RequiredArtifacts: RequiredArtifact list }
@@ -77,8 +88,12 @@ type State =

 // 6. Contract Signatures

+val parseBundleLocation : TaintedBundleInput -> Result<TrustedBundleLocation, Error>
+
 val deriveRunIdentity : SelectedSnapshot -> Result<DerivedRunIdentity, Error>

+val validatePreviousRunManifest : RunManifest -> Result<VerifiedPreviousRunManifest, Error>
+
 val validateSnapshotSelection : State -> IngestUpstreamSnapshot -> Result<SnapshotReady, Error>

 val decideSegmentRecords : SnapshotReady -> Result<DeterministicSegmentsReady, Error>
@@ -4,7 +4,9 @@ module IngestSnapshot.SharedModel

 type SnapshotIdentity = SnapshotIdentity of string

-type BundleSource = BundleSource of string
+type TaintedBundleLocation = TaintedBundleLocation of string
+
+type TrustedBundleLocation = TrustedBundleLocation of string

 type RunIdentity = RunIdentity of string

@@ -20,13 +22,13 @@ type NormalizedHash = NormalizedHash of string

 type ShapeHash = ShapeHash of string

-type ManifestPath = ManifestPath of string
+type TrustedManifestPath = TrustedManifestPath of string

-type SegmentsPath = SegmentsPath of string
+type TrustedSegmentsPath = TrustedSegmentsPath of string

-type CanonicalProjectionPath = CanonicalProjectionPath of string
+type TrustedCanonicalProjectionPath = TrustedCanonicalProjectionPath of string

-type SummaryPath = SummaryPath of string
+type TrustedSummaryPath = TrustedSummaryPath of string

 // 2. Shared compounds

@@ -36,9 +38,11 @@ type SnapshotMetadata =

 type SelectedSnapshot =
    { SnapshotIdentity: SnapshotIdentity
-      BundleSource: BundleSource
+      BundleLocation: TrustedBundleLocation
      SnapshotMetadata: SnapshotMetadata option }

+type VerifiedPreviousRunManifest = VerifiedPreviousRunManifest of RunManifest
+
 type SegmentHashes =
    { RawHash: RawHash
      NormalizedHash: NormalizedHash
@@ -54,7 +58,7 @@ type SegmentRecord =
 type RunManifest =
    { RunIdentity: RunIdentity
      SnapshotIdentity: SnapshotIdentity
-      ManifestPath: ManifestPath
-      SegmentsPath: SegmentsPath
-      CanonicalProjectionPath: CanonicalProjectionPath
-      SummaryPath: SummaryPath option }
+      ManifestPath: TrustedManifestPath
+      SegmentsPath: TrustedSegmentsPath
+      CanonicalProjectionPath: TrustedCanonicalProjectionPath
+      SummaryPath: TrustedSummaryPath option }