feat: sh-openswe-traces — LangSmith bulk-export trace archive

Storage-only SAM stack (S3 + KMS CMK + write-only IAM writer + Secrets Manager holder) as the S3 destination for LangSmith Bulk Export of Open SWE traces, for long-horizon auditing and prompt improvement (Athena over Parquet). - template.yaml: versioned SSE-KMS bucket, access-log bucket, TLS-only policy, DEEP_ARCHIVE lifecycle; least-privilege LangSmith writer (bucket-wide PutObject, ViaService-scoped KMS, no read/delete). - bootstrap.yaml: dedicated OIDC deploy role + least-privilege CFN exec role so CI never touches the shared execution role. - CI/CD via reusable ci-python-sam / cd-sam workflows. IAM passed GPT-4.1 cross-review + /sh-security-review (no blocking findings).
2026-09-30 04:33:14 +00:00 · 2026-07-10 15:45:23 -04:00 · 2026-07-10 15:45:23 -04:00 · 0dea585c1a
commit 0dea585c1a
7 changed files with 682 additions and 0 deletions
--- a/.github/workflows/ci.yaml
+++ b/.github/workflows/ci.yaml
@ -0,0 +1,10 @@
+name: CI
+on:
+  pull_request:
+    branches: [main]
+
+jobs:
+  ci:
+    # No Python source in this repo — ruff no-ops on an empty file set and the
+    # reusable workflow still runs `sam validate --lint` on template.yaml.
+    uses: Sea-Haven-Industries/.github/.github/workflows/ci-python-sam.yaml@fd60e4c9041784f666ac0fdefb9bec3c7fbf5143 # main
--- a/.github/workflows/deploy.yaml
+++ b/.github/workflows/deploy.yaml
@ -0,0 +1,22 @@
+name: Deploy
+on:
+  push:
+    branches: [main]
+
+permissions:
+  id-token: write
+  contents: read
+
+concurrency:
+  group: deploy
+  cancel-in-progress: false
+
+jobs:
+  deploy:
+    uses: Sea-Haven-Industries/.github/.github/workflows/cd-sam.yaml@fd60e4c9041784f666ac0fdefb9bec3c7fbf5143 # main
+    with:
+      stack-name: sh-openswe-traces
+      cfn-role-arn: arn:aws:iam::328440206208:role/sh-openswe-traces-cfn-exec-role
+    secrets:
+      deploy-role-arn: ${{ secrets.AWS_DEPLOY_ROLE_ARN }}
+      parameter-overrides: ${{ secrets.SAM_PARAMETER_OVERRIDES }}
--- a/.gitignore
+++ b/.gitignore
@ -0,0 +1,5 @@
+.aws-sam/
+samconfig.toml
+*.pyc
+__pycache__/
+.DS_Store
--- a/README.md
+++ b/README.md
@ -0,0 +1,194 @@
+# sh-openswe-traces
+
+LangSmith Bulk Export destination for the Open SWE deployment. A **storage-only**
+SAM stack — no compute. LangSmith runs the export on its own schedule and writes
+Parquet run/trace data into this bucket; we retain it for periodic auditing and
+prompt/instruction improvement (query with Athena).
+
+## Why this exists
+
+LangSmith retains traces for ~14 days. To audit agent behavior over longer
+horizons and mine it for prompt improvements, we own the data in S3 rather than
+paying for extended LangSmith retention. Capture is LangSmith-native Bulk Export;
+this repo is just the destination and its access controls.
+
+## Architecture
+
+```
+LangSmith (Bulk Export, scheduled LangSmith-side)
+        │  s3:PutObject  (IAM user access key, least-privilege)
+        ▼
+  s3://sh-openswe-traces/langsmith/…    SSE-KMS (alias/sh-openswe-traces)
+        │  lifecycle: → DEEP_ARCHIVE @ 90d
+        ▼
+  Athena / manual audit
+```
+
+| Resource | Name | Notes |
+|---|---|---|
+| S3 bucket | `sh-openswe-traces` | BPA all-on, SSE-KMS (default), versioned, TLS-only, `Retain` on delete |
+| Log bucket | `sh-openswe-traces-logs` | S3 server access logs (SSE-S3) for read attribution; logs expire 365d |
+| KMS CMK | `alias/sh-openswe-traces` | Rotation on; bucket default + writer encrypt through it |
+| IAM user | auto-named (tag `sh-openswe-langsmith-export`) | Write-only LangSmith writer; `PutObject` bucket-wide, no read/delete |
+| Secret | `sh-openswe/langsmith-export-s3` | Writer's access key (aws/secretsmanager key); populated post-deploy |
+
+The IAM user is **not** given an explicit name so the app stack deploys under
+`CAPABILITY_IAM`. It is referenced by ARN.
+
+## Deploy
+
+Two stacks:
+
+- **`bootstrap.yaml`** → `sh-openswe-traces-bootstrap` — the CI IAM roles (OIDC deploy
+  role + a least-privilege CFN exec role). Deployed **once, manually, under admin**
+  (`CAPABILITY_NAMED_IAM`); rarely changes. Kept separate so CI never touches the shared
+  `github-cfn-execution-role`, which is roles-only and can't create this stack's KMS key,
+  secret, or IAM user.
+- **`template.yaml`** → `sh-openswe-traces` — the app (buckets, CMK, writer, secret). CI
+  (`cd-sam.yaml`) deploys it on merge to `main`, assuming the OIDC deploy role and passing
+  `sh-openswe-traces-cfn-exec-role` as `--role-arn`.
+
+```bash
+# One-time bootstrap (admin):
+aws cloudformation deploy --template-file bootstrap.yaml \
+  --stack-name sh-openswe-traces-bootstrap --capabilities CAPABILITY_NAMED_IAM \
+  --region us-east-1
+
+# App stack — CI does this on merge to main; for a local/admin deploy:
+cp samconfig.toml.example samconfig.toml
+sam validate --lint && sam build
+sam deploy --capabilities CAPABILITY_IAM --resolve-s3
+```
+
+### Wire CI (after bootstrap + repo exist)
+
+- Set repo secret **`AWS_DEPLOY_ROLE_ARN`** to the bootstrap `DeployRoleArn` output
+  (`arn:aws:iam::328440206208:role/githubdeploy-sh-openswe-traces`).
+- `deploy.yaml` already passes `cfn-role-arn = sh-openswe-traces-cfn-exec-role`; the OIDC
+  trust is pinned to this repo's `main` ref.
+
+### Post-deploy: mint and store the writer access key
+
+The stack creates the IAM user and an empty secret; the access key is minted
+out-of-band so it never lands in CloudFormation state. **Run this only on a trusted
+single-user workstation, never in CI** — it handles a live credential.
+
+```bash
+USER=$(aws cloudformation describe-stacks --stack-name sh-openswe-traces \
+  --query "Stacks[0].Outputs[?OutputKey=='ExportUserName'].OutputValue" --output text)
+
+KEY_JSON=$(aws iam create-access-key --user-name "$USER" \
+  | jq '{AccessKeyId: .AccessKey.AccessKeyId, SecretAccessKey: .AccessKey.SecretAccessKey}')
+
+# Pass the secret via stdin, not argv, so it never appears in the process table.
+aws secretsmanager put-secret-value \
+  --secret-id sh-openswe/langsmith-export-s3 \
+  --secret-string file:///dev/stdin <<<"$KEY_JSON"
+
+unset KEY_JSON
+```
+
+### Configure LangSmith Bulk Export
+
+Driven by the LangSmith API (Plus/Enterprise only). Needs `LS_API_KEY` (LangSmith API
+key) and `LS_TENANT` (workspace id). Run **after** the writer key is minted/stored — the
+destination call validates by test-writing to the bucket.
+
+**1. Create the destination** (creds pulled from Secrets Manager, never pasted):
+
+```bash
+CREDS=$(aws secretsmanager get-secret-value --secret-id sh-openswe/langsmith-export-s3 \
+  --query SecretString --output text)
+AKID=$(jq -r .AccessKeyId <<<"$CREDS"); SAK=$(jq -r .SecretAccessKey <<<"$CREDS")
+
+curl -sS -X POST 'https://api.smith.langchain.com/api/v1/bulk-exports/destinations' \
+  -H 'Content-Type: application/json' -H "X-API-Key: $LS_API_KEY" -H "X-Tenant-Id: $LS_TENANT" \
+  --data @- <<JSON | jq .
+{ "destination_type": "s3", "display_name": "sh-openswe-traces us-east-1",
+  "config": { "bucket_name": "sh-openswe-traces", "prefix": "langsmith", "region": "us-east-1" },
+  "credentials": { "access_key_id": "$AKID", "secret_access_key": "$SAK" } }
+JSON
+```
+
+`display_name` must match `^[a-zA-Z0-9\-_ ']+$` (no parens); omit `endpoint_url` (S3-native,
+not GCS/MinIO). Save the returned destination `id`. LangSmith writes header-less, so
+bucket-default SSE-KMS encrypts every object under the CMK (confirm via `head-object`).
+
+**2. One scheduled export per tracing project** (this workspace has 4). Get project UUIDs
+from `GET /api/v1/sessions`, then:
+
+```bash
+export LS_DEST_ID='<destination id>'
+PROJECTS=( '<uuid-1>' '<uuid-2>' '<uuid-3>' '<uuid-4>' )   # all 4, or the subset you audit
+for PID in "${PROJECTS[@]}"; do
+  curl -sS -X POST 'https://api.smith.langchain.com/api/v1/bulk-exports' \
+    -H 'Content-Type: application/json' -H "X-API-Key: $LS_API_KEY" -H "X-Tenant-Id: $LS_TENANT" \
+    --data @- <<JSON | jq '{id, session_id, status}'
+{ "bulk_export_destination_id": "$LS_DEST_ID", "session_id": "$PID",
+  "start_time": "2026-06-26T00:00:00Z", "interval_hours": 24, "format_version": "v2_beta" }
+JSON
+done
+```
+
+`start_time` ~14d back backfills each project's retained window, then it continues daily.
+Keep `inputs`/`outputs` (omit `export_fields`) — they're the point of the audit. Data lands
+partitioned per project: `langsmith/export_id=…/…/session_id=<id>/…`.
+
+### Monitor exports
+
+```bash
+# Every export in the workspace: id, project, schedule, status
+curl -sS 'https://api.smith.langchain.com/api/v1/bulk-exports' \
+  -H "X-API-Key: $LS_API_KEY" -H "X-Tenant-Id: $LS_TENANT" \
+  | jq -r '(.bulk_exports // .exports // .)[]
+           | [.id,
+              (.session_id // "all_experiments"),
+              (if .interval_hours then "every \(.interval_hours)h" else "one-off" end),
+              .status] | @tsv' | column -t
+```
+
+- A recurring schedule's own status is `IntervalScheduled`; the daily child exports it
+  spawns have `CREATED` / `RUNNING` / `COMPLETED` / `FAILED` / `CANCELLED` / `TIMEDOUT`
+  (child exports carry `source_bulk_export_id`).
+- One export's detail: `GET /api/v1/bulk-exports/<id>` — its per-run rows:
+  `GET /api/v1/bulk-exports/<id>/runs`.
+- **Stop a schedule:** `PATCH /api/v1/bulk-exports/<id>` with `{"status":"Cancelled"}`.
+  Already-spawned child exports must be cancelled separately, and a cancelled job can't be
+  restarted — create a new one.
+
+## Operations
+
+- **Rotate** the writer access key quarterly: `aws iam create-access-key`, update the
+  secret + LangSmith destination (`PATCH`/recreate), then delete the old key. Key age is
+  monitored account-wide by the Security Hub `ACCESS_KEYS_ROTATED` Config rule (flags at
+  90d) — rotation keeps it compliant.
+- **Audit**: point Athena at `s3://sh-openswe-traces/langsmith/` (Parquet). Objects older
+  than 90 days are in Deep Archive — restore before querying.
+- **Read attribution**: object access is logged to `s3://sh-openswe-traces-logs/s3-access/`
+  (S3 server access logging).
+- **Cost**: Deep Archive ≈ $1/TB/mo; expect the archive to dominate storage cost.
+
+## Gotchas (from IAM cross-review)
+
+- **Writer needs `kms:Decrypt`.** SSE-KMS multipart uploads call `kms:Decrypt` at
+  `CompleteMultipartUpload`; without it every multipart export fails `AccessDenied`.
+  It's granted and safe — the writer has no `s3:GetObject`, so nothing to exfiltrate.
+- **No ACL headers.** The bucket is `BucketOwnerEnforced`; any PutObject carrying an
+  ACL header is rejected with `AccessControlListNotSupported` (not fixable in policy).
+  boto3/most SDKs send none by default — verify LangSmith's exporter likewise.
+- **Downstream readers need their own CMK grant.** An Athena/Glue role reading the
+  Parquet needs explicit `kms:Decrypt` (+ `kms:GenerateDataKey`) on
+  `alias/sh-openswe-traces` — see `reference_kms_cmk_grant_migration`.
+
+## Security
+
+Traces can contain source code and secrets surfaced in tool I/O. Controls: SSE-KMS at
+rest (customer-managed CMK, bucket default), versioning (overwrite recovery), Block Public
+Access, TLS-only bucket policy, a write-only least-privilege writer (bucket-wide `PutObject`
+ `kms:Decrypt` gated to `kms:ViaService=s3`, no read/delete), the credential secret on a
+separate managed key, and S3 access logging. CI deploys via a dedicated least-privilege
+exec role (see `bootstrap.yaml`), not the shared execution role.
+
+The IAM surface passed GPT-4.1 cross-review and a `/sh-security-review` fan-out +
+proof-or-kill verifier (no blocking findings). Deferred, non-blocking: CloudTrail S3
+data-events (org trail carries none; access logging covers attribution for now).
--- a/bootstrap.yaml
+++ b/bootstrap.yaml
@ -0,0 +1,202 @@
+AWSTemplateFormatVersion: "2010-09-09"
+Description: >
+  Bootstrap IAM for the sh-openswe-traces app stack. Deployed ONCE, manually, under
+  admin (CAPABILITY_NAMED_IAM). Creates two named roles so CI never touches the shared
+  github-cfn-execution-role:
+    - DeployRole (githubdeploy-sh-openswe-traces): assumed by this repo's GitHub
+      Actions via OIDC (main branch only); can drive CloudFormation for THIS stack and
+      pass the exec role.
+    - ExecRole (sh-openswe-traces-cfn-exec-role): assumed by CloudFormation to create
+      the app stack's resources; least-privilege to exactly this stack's resource set.
+
+Resources:
+  ExecRole:
+    Type: AWS::IAM::Role
+    Properties:
+      RoleName: sh-openswe-traces-cfn-exec-role
+      Description: CloudFormation execution role for the sh-openswe-traces app stack.
+      AssumeRolePolicyDocument:
+        Version: "2012-10-17"
+        Statement:
+          - Effect: Allow
+            Principal:
+              Service: cloudformation.amazonaws.com
+            Action: sts:AssumeRole
+            Condition:
+              StringEquals:
+                "aws:SourceAccount": !Ref "AWS::AccountId"
+      Policies:
+        - PolicyName: manage-sh-openswe-traces-resources
+          PolicyDocument:
+            Version: "2012-10-17"
+            Statement:
+              # Defense-in-depth: this role can create sh-openswe-traces-* users, so
+              # explicitly forbid the actions that would turn one into a usable/escalated
+              # principal (credentials, console login, extra policies, boundary removal).
+              - Sid: DenyUserCredentialAndEscalation
+                Effect: Deny
+                Action:
+                  - "iam:CreateAccessKey"
+                  - "iam:CreateLoginProfile"
+                  - "iam:UpdateLoginProfile"
+                  - "iam:AttachUserPolicy"
+                  - "iam:CreateServiceSpecificCredential"
+                  - "iam:PutUserPermissionsBoundary"
+                  - "iam:DeleteUserPermissionsBoundary"
+                Resource: "*"
+              - Sid: Buckets
+                Effect: Allow
+                Action:
+                  - "s3:CreateBucket"
+                  - "s3:DeleteBucket"
+                  - "s3:PutBucketPolicy"
+                  - "s3:DeleteBucketPolicy"
+                  - "s3:GetBucketPolicy"
+                  - "s3:PutEncryptionConfiguration"
+                  - "s3:GetEncryptionConfiguration"
+                  - "s3:PutBucketVersioning"
+                  - "s3:GetBucketVersioning"
+                  - "s3:PutBucketPublicAccessBlock"
+                  - "s3:GetBucketPublicAccessBlock"
+                  - "s3:PutBucketOwnershipControls"
+                  - "s3:GetBucketOwnershipControls"
+                  - "s3:PutLifecycleConfiguration"
+                  - "s3:GetLifecycleConfiguration"
+                  - "s3:PutBucketLogging"
+                  - "s3:GetBucketLogging"
+                  - "s3:PutBucketTagging"
+                  - "s3:GetBucketTagging"
+                  - "s3:GetBucketLocation"
+                  - "s3:GetBucketAcl"
+                Resource:
+                  - "arn:aws:s3:::sh-openswe-traces"
+                  - "arn:aws:s3:::sh-openswe-traces-logs"
+              # CreateKey/CreateAlias cannot be resource-scoped (the key does not yet
+              # exist). Only CloudFormation can assume this role, and only to deploy this
+              # stack, so the blast radius is bounded to this stack's deployments.
+              - Sid: Kms
+                Effect: Allow
+                Action:
+                  - "kms:CreateKey"
+                  - "kms:CreateAlias"
+                  - "kms:DeleteAlias"
+                  - "kms:UpdateAlias"
+                  - "kms:PutKeyPolicy"
+                  - "kms:GetKeyPolicy"
+                  - "kms:EnableKeyRotation"
+                  - "kms:DisableKeyRotation"
+                  - "kms:GetKeyRotationStatus"
+                  - "kms:DescribeKey"
+                  - "kms:TagResource"
+                  - "kms:UntagResource"
+                  - "kms:ListResourceTags"
+                  - "kms:ScheduleKeyDeletion"
+                  - "kms:EnableKey"
+                Resource: "*"
+              - Sid: Secret
+                Effect: Allow
+                Action:
+                  - "secretsmanager:CreateSecret"
+                  - "secretsmanager:DeleteSecret"
+                  - "secretsmanager:DescribeSecret"
+                  - "secretsmanager:UpdateSecret"
+                  - "secretsmanager:TagResource"
+                  - "secretsmanager:UntagResource"
+                  - "secretsmanager:GetResourcePolicy"
+                  - "secretsmanager:PutResourcePolicy"
+                Resource: !Sub "arn:aws:secretsmanager:${AWS::Region}:${AWS::AccountId}:secret:sh-openswe/*"
+              - Sid: ExportUser
+                Effect: Allow
+                Action:
+                  - "iam:CreateUser"
+                  - "iam:DeleteUser"
+                  - "iam:GetUser"
+                  - "iam:TagUser"
+                  - "iam:UntagUser"
+                  - "iam:PutUserPolicy"
+                  - "iam:DeleteUserPolicy"
+                  - "iam:GetUserPolicy"
+                  - "iam:ListUserPolicies"
+                  - "iam:ListUserTags"
+                  - "iam:ListAttachedUserPolicies"
+                  - "iam:ListGroupsForUser"
+                Resource: !Sub "arn:aws:iam::${AWS::AccountId}:user/sh-openswe-traces-*"
+
+  DeployRole:
+    Type: AWS::IAM::Role
+    Properties:
+      RoleName: githubdeploy-sh-openswe-traces
+      Description: GitHub Actions OIDC deploy role for the sh-openswe-traces app stack.
+      MaxSessionDuration: 3600
+      AssumeRolePolicyDocument:
+        Version: "2012-10-17"
+        Statement:
+          - Effect: Allow
+            Principal:
+              Federated: !Sub "arn:aws:iam::${AWS::AccountId}:oidc-provider/token.actions.githubusercontent.com"
+            Action: sts:AssumeRoleWithWebIdentity
+            Condition:
+              StringEquals:
+                "token.actions.githubusercontent.com:aud": "sts.amazonaws.com"
+                "token.actions.githubusercontent.com:sub": "repo:Sea-Haven-Industries/sh-openswe-traces:ref:refs/heads/main"
+      Policies:
+        - PolicyName: deploy-sh-openswe-traces
+          PolicyDocument:
+            Version: "2012-10-17"
+            Statement:
+              - Sid: AppStack
+                Effect: Allow
+                Action:
+                  - "cloudformation:CreateChangeSet"
+                  - "cloudformation:ExecuteChangeSet"
+                  - "cloudformation:DescribeChangeSet"
+                  - "cloudformation:DeleteChangeSet"
+                  - "cloudformation:CreateStack"
+                  - "cloudformation:UpdateStack"
+                  - "cloudformation:DescribeStacks"
+                  - "cloudformation:DescribeStackEvents"
+                  - "cloudformation:DescribeStackResource"
+                  - "cloudformation:DescribeStackResources"
+                  - "cloudformation:ListStackResources"
+                  - "cloudformation:GetTemplate"
+                  - "cloudformation:GetTemplateSummary"
+                Resource:
+                  - !Sub "arn:aws:cloudformation:${AWS::Region}:${AWS::AccountId}:stack/sh-openswe-traces/*"
+                  - !Sub "arn:aws:cloudformation:${AWS::Region}:${AWS::AccountId}:changeSet/*/*"
+              # SAM's --resolve-s3 looks up (does not recreate) the pre-existing managed
+              # artifact stack + bucket.
+              - Sid: SamManagedStackRead
+                Effect: Allow
+                Action:
+                  - "cloudformation:DescribeStacks"
+                Resource: !Sub "arn:aws:cloudformation:${AWS::Region}:${AWS::AccountId}:stack/aws-sam-cli-managed-default/*"
+              - Sid: SamArtifactBucket
+                Effect: Allow
+                Action:
+                  - "s3:GetObject"
+                  - "s3:PutObject"
+                  - "s3:GetBucketLocation"
+                  - "s3:ListBucket"
+                Resource:
+                  - "arn:aws:s3:::aws-sam-cli-managed-default-*"
+                  - "arn:aws:s3:::aws-sam-cli-managed-default-*/*"
+              - Sid: CfnValidate
+                Effect: Allow
+                Action:
+                  - "cloudformation:ValidateTemplate"
+                Resource: "*"
+              - Sid: PassExecRoleToCfn
+                Effect: Allow
+                Action: "iam:PassRole"
+                Resource: !GetAtt ExecRole.Arn
+                Condition:
+                  StringEquals:
+                    "iam:PassedToService": cloudformation.amazonaws.com
+
+Outputs:
+  DeployRoleArn:
+    Description: Set as the repo secret AWS_DEPLOY_ROLE_ARN.
+    Value: !GetAtt DeployRole.Arn
+  ExecRoleArn:
+    Description: Set as cfn-role-arn in .github/workflows/deploy.yaml.
+    Value: !GetAtt ExecRole.Arn
--- a/samconfig.toml.example
+++ b/samconfig.toml.example
@ -0,0 +1,10 @@
+version = 0.1
+
+[default.deploy.parameters]
+stack_name = "sh-openswe-traces"
+region = "us-east-1"
+capabilities = "CAPABILITY_IAM"
+resolve_s3 = true
+confirm_changeset = true
+# Optional: override the export prefix
+# parameter_overrides = "ExportPrefix=langsmith/"
--- a/template.yaml
+++ b/template.yaml
@ -0,0 +1,239 @@
+AWSTemplateFormatVersion: "2010-09-09"
+Transform: AWS::Serverless-2016-10-31
+Description: >
+  sh-openswe-traces — LangSmith Bulk Export destination. Storage-only:
+  KMS-encrypted S3 bucket, a least-privilege IAM writer for LangSmith's
+  export job, and a Secrets Manager holder for that writer's access key.
+  No compute — the export schedule is configured on the LangSmith side.
+
+Parameters:
+  ExportPrefix:
+    Type: String
+    Default: langsmith/
+    Description: S3 key prefix LangSmith writes exports under (also the lifecycle scope).
+
+Resources:
+  TracesKey:
+    Type: AWS::KMS::Key
+    Properties:
+      Description: SSE-KMS CMK for sh-openswe-traces (LangSmith export archive).
+      EnableKeyRotation: true
+      KeyPolicy:
+        Version: "2012-10-17"
+        Statement:
+          # Root-enable so IAM identity policies (below) govern access.
+          - Sid: EnableIAMPolicies
+            Effect: Allow
+            Principal:
+              AWS: !Sub "arn:aws:iam::${AWS::AccountId}:root"
+            Action: "kms:*"
+            Resource: "*"
+
+  TracesKeyAlias:
+    Type: AWS::KMS::Alias
+    Properties:
+      AliasName: alias/sh-openswe-traces
+      TargetKeyId: !Ref TracesKey
+
+  TracesBucket:
+    Type: AWS::S3::Bucket
+    DeletionPolicy: Retain
+    UpdateReplacePolicy: Retain
+    # Logging target policy must exist before S3 will accept LoggingConfiguration.
+    DependsOn: TracesLogBucketPolicy
+    Properties:
+      BucketName: sh-openswe-traces
+      PublicAccessBlockConfiguration:
+        BlockPublicAcls: true
+        BlockPublicPolicy: true
+        IgnorePublicAcls: true
+        RestrictPublicBuckets: true
+      BucketEncryption:
+        ServerSideEncryptionConfiguration:
+          - ServerSideEncryptionByDefault:
+              SSEAlgorithm: aws:kms
+              KMSMasterKeyID: !Ref TracesKey
+            BucketKeyEnabled: true
+      OwnershipControls:
+        Rules:
+          - ObjectOwnership: BucketOwnerEnforced
+      # Tamper recovery: the writer key is write-only (no DeleteObject / no
+      # DeleteObjectVersion), so a malicious or buggy overwrite creates a noncurrent
+      # version the prior bytes are recoverable from. Exports write new partitioned
+      # keys, so noncurrent versions are rare — expire them after 90 days.
+      VersioningConfiguration:
+        Status: Enabled
+      LifecycleConfiguration:
+        Rules:
+          - Id: archive-exports-to-deep-archive
+            Status: Enabled
+            Prefix: !Ref ExportPrefix
+            Transitions:
+              - StorageClass: DEEP_ARCHIVE
+                TransitionInDays: 90
+          - Id: expire-noncurrent-versions
+            Status: Enabled
+            NoncurrentVersionExpiration:
+              NoncurrentDays: 90
+          - Id: abort-incomplete-multipart
+            Status: Enabled
+            AbortIncompleteMultipartUpload:
+              DaysAfterInitiation: 7
+      LoggingConfiguration:
+        DestinationBucketName: !Ref TracesLogBucket
+        LogFilePrefix: s3-access/
+
+  TracesBucketPolicy:
+    Type: AWS::S3::BucketPolicy
+    Properties:
+      Bucket: !Ref TracesBucket
+      PolicyDocument:
+        Version: "2012-10-17"
+        Statement:
+          - Sid: DenyInsecureTransport
+            Effect: Deny
+            Principal: "*"
+            Action: "s3:*"
+            Resource:
+              - !GetAtt TracesBucket.Arn
+              - !Sub "${TracesBucket.Arn}/*"
+            Condition:
+              Bool:
+                "aws:SecureTransport": "false"
+          # No SSE-header enforcement Deny. LangSmith's exporter does not send an
+          # "aws:kms" SSE header, so a "must be aws:kms" Deny blocks its writes — and
+          # StringNotEqualsIfExists on a Deny also blocks header-less puts (absent key
+          # evaluates true). Encryption is instead guaranteed by the bucket DEFAULT
+          # (SSE-KMS with our CMK, applied to every header-less put) plus S3's baseline
+          # (no object is ever stored unencrypted). If LangSmith explicitly requests
+          # AES256, that object lands as SSE-S3 rather than CMK — verify post-write
+          # (head-object) and decide CMK-vs-SSE-S3 if so.
+
+  # Server access logging target for TracesBucket — read attribution for the
+  # secret-bearing archive (the org trail logs no S3 data events). SSE-S3 only:
+  # S3 log delivery cannot write to an SSE-KMS bucket.
+  TracesLogBucket:
+    Type: AWS::S3::Bucket
+    DeletionPolicy: Retain
+    UpdateReplacePolicy: Retain
+    Properties:
+      BucketName: sh-openswe-traces-logs
+      PublicAccessBlockConfiguration:
+        BlockPublicAcls: true
+        BlockPublicPolicy: true
+        IgnorePublicAcls: true
+        RestrictPublicBuckets: true
+      BucketEncryption:
+        ServerSideEncryptionConfiguration:
+          - ServerSideEncryptionByDefault:
+              SSEAlgorithm: AES256
+      OwnershipControls:
+        Rules:
+          - ObjectOwnership: BucketOwnerEnforced
+      LifecycleConfiguration:
+        Rules:
+          - Id: expire-access-logs
+            Status: Enabled
+            ExpirationInDays: 365
+
+  TracesLogBucketPolicy:
+    Type: AWS::S3::BucketPolicy
+    Properties:
+      Bucket: !Ref TracesLogBucket
+      PolicyDocument:
+        Version: "2012-10-17"
+        Statement:
+          - Sid: S3ServerAccessLogsWrite
+            Effect: Allow
+            Principal:
+              Service: logging.s3.amazonaws.com
+            Action: "s3:PutObject"
+            Resource: !Sub "${TracesLogBucket.Arn}/s3-access/*"
+            Condition:
+              ArnLike:
+                # literal (not !GetAtt) to avoid a cycle with TracesBucket's DependsOn
+                "aws:SourceArn": "arn:aws:s3:::sh-openswe-traces"
+              StringEquals:
+                "aws:SourceAccount": !Ref "AWS::AccountId"
+          - Sid: DenyInsecureTransport
+            Effect: Deny
+            Principal: "*"
+            Action: "s3:*"
+            Resource:
+              - !GetAtt TracesLogBucket.Arn
+              - !Sub "${TracesLogBucket.Arn}/*"
+            Condition:
+              Bool:
+                "aws:SecureTransport": "false"
+
+  # No explicit UserName: an IAM name would require CAPABILITY_NAMED_IAM, but the
+  # standard cd-sam.yaml reusable workflow deploys with CAPABILITY_IAM only. The
+  # principal is referenced by ARN (in the secret + LangSmith config), not by name;
+  # the tag carries the human-facing identifier.
+  LangSmithExportUser:
+    Type: AWS::IAM::User
+    Properties:
+      Tags:
+        - Key: Name
+          Value: sh-openswe-langsmith-export
+        - Key: purpose
+          Value: langsmith-bulk-export-writer
+      Policies:
+        - PolicyName: langsmith-export-put
+          PolicyDocument:
+            Version: "2012-10-17"
+            Statement:
+              # Bucket-wide (not prefix-scoped): LangSmith's destination-creation
+              # validation writes a test object whose key is NOT guaranteed to be
+              # under ExportPrefix (docs reference a /tmp path), and its documented
+              # policy scopes PutObject to the whole bucket. This bucket is
+              # single-purpose, so bucket-wide write is still tightly bounded.
+              # Deliberately NO s3:GetObject / s3:DeleteObject (both optional per
+              # LangSmith): omitting them keeps the writer write-only (no exfil,
+              # no delete). Trade-off: LangSmith skips post-write size verification
+              # and leaves its small test object behind (harmless).
+              - Sid: PutExportObjects
+                Effect: Allow
+                Action:
+                  - "s3:PutObject"
+                  - "s3:AbortMultipartUpload"
+                Resource: !Sub "${TracesBucket.Arn}/*"
+              - Sid: EncryptWithBucketKey
+                Effect: Allow
+                Action:
+                  - "kms:GenerateDataKey"
+                  - "kms:Encrypt"
+                  # Required by S3 at CompleteMultipartUpload for SSE-KMS. Safe: the
+                  # writer has no s3:GetObject, so there is no object to decrypt/exfil.
+                  - "kms:Decrypt"
+                Resource: !GetAtt TracesKey.Arn
+                # Usable only through S3 — blocks a leaked key from calling kms:Decrypt
+                # directly against arbitrary ciphertext under this CMK.
+                Condition:
+                  StringEquals:
+                    "kms:ViaService": !Sub "s3.${AWS::Region}.amazonaws.com"
+
+  # Holder only — the real access key is minted post-deploy and written in
+  # with `aws secretsmanager put-secret-value` (see README). Never in the template.
+  ExportKeySecret:
+    Type: AWS::SecretsManager::Secret
+    Properties:
+      Name: sh-openswe/langsmith-export-s3
+      Description: >
+        Access key for the sh-openswe-langsmith-export IAM user, consumed by
+        LangSmith Bulk Export. Populated out-of-band post-deploy; rotate quarterly.
+      # Encrypted with the aws/secretsmanager managed key — deliberately NOT the
+      # trace CMK, so the credential and the data it protects never share a key the
+      # writer principal holds any KMS grant on.
+
+Outputs:
+  BucketName:
+    Value: !Ref TracesBucket
+  BucketArn:
+    Value: !GetAtt TracesBucket.Arn
+  KmsKeyArn:
+    Value: !GetAtt TracesKey.Arn
+  ExportUserName:
+    Value: !Ref LangSmithExportUser
+  ExportKeySecretName:
+    Value: sh-openswe/langsmith-export-s3