Skip to content

chore: Wave 9 engineering cleanup and Wave E/11/13 delivery - #229

Merged
fraware merged 36 commits into
mainfrom
chore/wave9-engineering-cleanup
Jul 23, 2026
Merged

chore: Wave 9 engineering cleanup and Wave E/11/13 delivery#229
fraware merged 36 commits into
mainfrom
chore/wave9-engineering-cleanup

Conversation

@fraware

@fraware fraware commented Jul 23, 2026

Copy link
Copy Markdown
Member

Summary

  • Harden DSSE fail-closed defaults, tool-broker tenant controls, ledger MCP rate limits, TS SDK retries, and policy-kernel Redis/DFA fixes.
  • Deliver Wave E launch wiring (Make/just/compose/pf-env), path-filtered CI with composite actions, Wave 11 cost cuts plus Windows smoke, and Wave 13 live DR/publish/revocation scripts.
  • Consolidate docs IA (getting-started/archive stubs) and purge unmaintained marketplace, ops terraform, on-ramps/recipes, and residual zombie tools.

Test plan

  • Confirm CI path filters and reusable Rust/Go/Node jobs still green on this PR
  • Spot-check DSSE fail-closed + TS SDK client tests
  • Verify docs site nav still builds (mkdocs) after archive moves
  • Ensure no secrets or build artifacts landed in the commits

fraware added 17 commits July 22, 2026 23:07
Require explicit allow-insecure opt-in so missing signatures cannot pass silently. Align Go, Rust, and TypeScript verifiers and cross-lang tests with deployment trust guidance.
Tone down overstated capabilities so the landing README matches what the repo actually builds and tests today.
Extend rate limiting with tenant-aware risk scoring, throttles, and budgets so broker abuse is contained without hard-coding a single global limit.
Protect the MCP proxy with counters and a sliding-window limiter so noisy clients cannot starve legitimate traffic.
Ship a retry-aware HTTP client, clearer errors, and drop unused gRPC deps so the TypeScript SDK matches how callers actually integrate.
Restore reliable Redis-backed caching and fix DFA compilation edge cases that could mis-label transitions under load.
Simplify export tooling so DFA and labeler generation stay maintainable and match the current Lean runtime APIs.
Centralize ENFORCED ExtendedAdapter paths and drop orphaned Policy/ProofBench roots so the Lean workspace has one clear entrypoint.
Keep egress, telemetry, JWKS/sidecar, and mpc-fintech crates aligned with the workspace Cargo graph and current runtime entrypoints.
Trim multiarch/criterion/paper spend, clarify SWE-bench expectations, and add a Windows smoke workflow so expensive or flaky gates fail soft with clear signals.
Add preflight/sync scripts and workflow wiring so cross-region DR, publish updates, and revocation sync can run with real secrets when available.
Unify Make/just/compose install paths with check_wiring and pf-env validation so local bring-up matches CI expectations.
Factor setup composites and reusable Rust/Go/Node jobs so PR CI only builds what changed and respects root npm workspaces.
Record latency budgets and remediation progress so Wave E/9+ work stays measurable against CI inventory and placeholder burn-down.
Re-home outdated feature and audit docs under archive stubs and sharpen the docs index so newcomers land on a single getting-started path.
Remove unmaintained marketplace, ops terraform, on-ramps/recipes/figures, and related CI/binaries so the tree reflects the supported product surface.
Delete one-off installers, duplicate VS Code extensions, and unused tooling while keeping adapter/evidence READMEs oriented to the current layout.
@cursor

cursor Bot commented Jul 23, 2026

Copy link
Copy Markdown

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

@github-actions

Copy link
Copy Markdown

Allowlist Sync Validation Failed

The runtime allowlist is out of sync with Lean proofs. Please run:

python3 tools/gen_allowlist_from_lean.py . runtime/sidecar-watcher/policy/allowlist.json

Then commit the updated allowlist to ensure runtime configuration matches formal specifications.

@github-actions

Copy link
Copy Markdown

SBOM Security Report

Total Packages: 3286

Language Distribution:

  • go: 323
  • javascript: 2003
  • rust: 654
  • python: 1

License Distribution:

  • MIT: 1319
  • Apache-2.0: 44
  • BSD-3-Clause: 43
  • CC0-1.0: 43
  • ISC: 84
  • BSD-2-Clause: 39
  • Python-2.0: 1
  • MPL-2.0: 1
  • Unlicense: 3
  • CC-BY-4.0: 1
  • (Apache-2.0 OR MPL-1.1): 1
  • BlueOak-1.0.0: 3
  • (AFL-2.1 OR BSD-3-Clause): 1
  • (BSD-3-Clause OR GPL-2.0): 1
  • (MIT AND Zlib): 1
  • (MIT AND BSD-3-Clause): 1
  • 0BSD: 4
  • (MIT OR CC0-1.0): 3
  • MIT AND ISC: 1

✅ SBOM generated successfully
✅ No GPL licenses detected
✅ No high/critical vulnerabilities found

@github-actions

Copy link
Copy Markdown

Sample Replay

@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

Repair multiarch format()/SC2193, bump crossbeam-epoch, declare crate licenses, fix mkdocs strict off-tree links, drop missing Cargo.lock COPY, and keep redteam sidecar build at repo root.
@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

The live docs/runbooks/rollback.md path was removed; link the archive GuardTrip triage page to runbook-rollback.md so mkdocs --strict stays honest.
@github-actions

Copy link
Copy Markdown

Allowlist Sync Validation Failed

The runtime allowlist is out of sync with Lean proofs. Please run:

python3 tools/gen_allowlist_from_lean.py . runtime/sidecar-watcher/policy/allowlist.json

Then commit the updated allowlist to ensure runtime configuration matches formal specifications.

@github-actions

Copy link
Copy Markdown

Sample Replay

@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

@github-actions

Copy link
Copy Markdown

Allowlist Sync Validation Failed

The runtime allowlist is out of sync with Lean proofs. Please run:

python3 tools/gen_allowlist_from_lean.py . runtime/sidecar-watcher/policy/allowlist.json

Then commit the updated allowlist to ensure runtime configuration matches formal specifications.

@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

@github-actions

Copy link
Copy Markdown

SBOM Security Report

Total Packages: 3286

Language Distribution:

  • go: 323
  • javascript: 2003
  • rust: 654
  • python: 1

License Distribution:

  • MIT: 1319
  • Apache-2.0: 44
  • BSD-3-Clause: 43
  • CC0-1.0: 43
  • ISC: 84
  • BSD-2-Clause: 39
  • Python-2.0: 1
  • MPL-2.0: 1
  • Unlicense: 3
  • CC-BY-4.0: 1
  • (Apache-2.0 OR MPL-1.1): 1
  • BlueOak-1.0.0: 3
  • (AFL-2.1 OR BSD-3-Clause): 1
  • (BSD-3-Clause OR GPL-2.0): 1
  • (MIT AND Zlib): 1
  • (MIT AND BSD-3-Clause): 1
  • 0BSD: 4
  • (MIT OR CC0-1.0): 3
  • MIT AND ISC: 1

✅ SBOM generated successfully
✅ No GPL licenses detected
✅ No high/critical vulnerabilities found

@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

Force real egress binary rebuild after dummy Cargo layer, bind 0.0.0.0:8081 with wget health tooling, add demo rate limits + log sanitization for CodeQL, and drop stale marketplace/ui install from CHANGELOG.
@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

@github-actions

Copy link
Copy Markdown

🔮 Stochastic Proof-Regression Results

Success Rate: 100.0% (minimum: 95%)
Trials: 50
Passed: 50
Failed: 0

Perturbation Results:

  • parameter_noise: 5/5 (100.0%)
  • requirement_addition: 10/10 (100.0%)
  • constraint_relaxation: 11/11 (100.0%)
  • requirement_removal: 6/6 (100.0%)
  • timeout_adjustment: 7/7 (100.0%)
  • constraint_tightening: 3/3 (100.0%)
  • metric_adjustment: 4/4 (100.0%)
  • resource_limit_adjustment: 4/4 (100.0%)

PASSED - Success rate meets minimum threshold.

View detailed results

@github-actions

Copy link
Copy Markdown

Sample Replay

@github-actions

Copy link
Copy Markdown

SBOM Security Report

Total Packages: 3286

Language Distribution:

  • go: 323
  • javascript: 2003
  • rust: 654
  • python: 1

License Distribution:

  • MIT: 1319
  • Apache-2.0: 44
  • BSD-3-Clause: 43
  • CC0-1.0: 43
  • ISC: 84
  • BSD-2-Clause: 39
  • Python-2.0: 1
  • MPL-2.0: 1
  • Unlicense: 3
  • CC-BY-4.0: 1
  • (Apache-2.0 OR MPL-1.1): 1
  • BlueOak-1.0.0: 3
  • (AFL-2.1 OR BSD-3-Clause): 1
  • (BSD-3-Clause OR GPL-2.0): 1
  • (MIT AND Zlib): 1
  • (MIT AND BSD-3-Clause): 1
  • 0BSD: 4
  • (MIT OR CC0-1.0): 3
  • MIT AND ISC: 1

✅ SBOM generated successfully
✅ No GPL licenses detected
✅ No high/critical vulnerabilities found

Comment thread runtime/ledger/minimal-server.js Fixed
Comment thread runtime/ledger/minimal-server.js Fixed
@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

Use express-rate-limit so CodeQL recognizes auth route limits, and
switch egress-firewall reqwest to rustls-tls to avoid musl OpenSSL
static linking failures in Docker/demo-e2e.
@github-actions

Copy link
Copy Markdown

🔮 Stochastic Proof-Regression Results

Success Rate: 100.0% (minimum: 95%)
Trials: 50
Passed: 50
Failed: 0

Perturbation Results:

  • parameter_noise: 5/5 (100.0%)
  • requirement_addition: 10/10 (100.0%)
  • requirement_removal: 6/6 (100.0%)
  • constraint_relaxation: 11/11 (100.0%)
  • timeout_adjustment: 7/7 (100.0%)
  • constraint_tightening: 3/3 (100.0%)
  • metric_adjustment: 4/4 (100.0%)
  • resource_limit_adjustment: 4/4 (100.0%)

PASSED - Success rate meets minimum threshold.

View detailed results

@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

@github-actions

Copy link
Copy Markdown

Sample Replay

@github-actions

Copy link
Copy Markdown

SBOM Security Report

Total Packages: 3288

Language Distribution:

  • go: 323
  • javascript: 2005
  • rust: 654
  • python: 1

License Distribution:

  • MIT: 1319
  • Apache-2.0: 44
  • BSD-3-Clause: 43
  • CC0-1.0: 43
  • ISC: 84
  • BSD-2-Clause: 39
  • Python-2.0: 1
  • MPL-2.0: 1
  • Unlicense: 3
  • CC-BY-4.0: 1
  • (Apache-2.0 OR MPL-1.1): 1
  • BlueOak-1.0.0: 3
  • (AFL-2.1 OR BSD-3-Clause): 1
  • (BSD-3-Clause OR GPL-2.0): 1
  • (MIT AND Zlib): 1
  • (MIT AND BSD-3-Clause): 1
  • 0BSD: 4
  • (MIT OR CC0-1.0): 3
  • MIT AND ISC: 1

✅ SBOM generated successfully
✅ No GPL licenses detected
✅ No high/critical vulnerabilities found

@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

wget --spider issues HEAD, which 404'd on a GET-only /health route and
marked the container unhealthy in demo-e2e. Accept HEAD and probe with GET.
@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

@github-actions

Copy link
Copy Markdown

🔮 Stochastic Proof-Regression Results

Success Rate: 100.0% (minimum: 95%)
Trials: 50
Passed: 50
Failed: 0

Perturbation Results:

  • constraint_relaxation: 11/11 (100.0%)
  • requirement_addition: 10/10 (100.0%)
  • requirement_removal: 6/6 (100.0%)
  • parameter_noise: 5/5 (100.0%)
  • timeout_adjustment: 7/7 (100.0%)
  • metric_adjustment: 4/4 (100.0%)
  • constraint_tightening: 3/3 (100.0%)
  • resource_limit_adjustment: 4/4 (100.0%)

PASSED - Success rate meets minimum threshold.

View detailed results

@github-actions

Copy link
Copy Markdown

Sample Replay

@github-actions

Copy link
Copy Markdown

SBOM Security Report

Total Packages: 3288

Language Distribution:

  • go: 323
  • javascript: 2005
  • rust: 654
  • python: 1

License Distribution:

  • MIT: 1319
  • Apache-2.0: 44
  • BSD-3-Clause: 43
  • CC0-1.0: 43
  • ISC: 84
  • BSD-2-Clause: 39
  • Python-2.0: 1
  • MPL-2.0: 1
  • Unlicense: 3
  • CC-BY-4.0: 1
  • (Apache-2.0 OR MPL-1.1): 1
  • BlueOak-1.0.0: 3
  • (AFL-2.1 OR BSD-3-Clause): 1
  • (BSD-3-Clause OR GPL-2.0): 1
  • (MIT AND Zlib): 1
  • (MIT AND BSD-3-Clause): 1
  • 0BSD: 4
  • (MIT OR CC0-1.0): 3
  • MIT AND ISC: 1

✅ SBOM generated successfully
✅ No GPL licenses detected
✅ No high/critical vulnerabilities found

@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

The CLI oneshot (wasm-sandbox --help) exits 0 and breaks
docker compose --profile full up --wait used by demo-up/demo-e2e.
Move it to the tools profile for compose run usage.
@github-actions

Copy link
Copy Markdown

Thanks for the PR! CI will run CERT validation and replay checks. Results will appear in workflow badges and artifacts.

@github-actions

Copy link
Copy Markdown

🔮 Stochastic Proof-Regression Results

Success Rate: 100.0% (minimum: 95%)
Trials: 50
Passed: 50
Failed: 0

Perturbation Results:

  • requirement_addition: 10/10 (100.0%)
  • constraint_relaxation: 11/11 (100.0%)
  • parameter_noise: 5/5 (100.0%)
  • requirement_removal: 6/6 (100.0%)
  • timeout_adjustment: 7/7 (100.0%)
  • constraint_tightening: 3/3 (100.0%)
  • metric_adjustment: 4/4 (100.0%)
  • resource_limit_adjustment: 4/4 (100.0%)

PASSED - Success rate meets minimum threshold.

View detailed results

@github-actions

Copy link
Copy Markdown

Sample Replay

@github-actions

Copy link
Copy Markdown

SBOM Security Report

Total Packages: 3288

Language Distribution:

  • go: 323
  • javascript: 2005
  • rust: 654
  • python: 1

License Distribution:

  • MIT: 1319
  • Apache-2.0: 44
  • BSD-3-Clause: 43
  • CC0-1.0: 43
  • ISC: 84
  • BSD-2-Clause: 39
  • Python-2.0: 1
  • MPL-2.0: 1
  • Unlicense: 3
  • CC-BY-4.0: 1
  • (Apache-2.0 OR MPL-1.1): 1
  • BlueOak-1.0.0: 3
  • (AFL-2.1 OR BSD-3-Clause): 1
  • (BSD-3-Clause OR GPL-2.0): 1
  • (MIT AND Zlib): 1
  • (MIT AND BSD-3-Clause): 1
  • 0BSD: 4
  • (MIT OR CC0-1.0): 3
  • MIT AND ISC: 1

✅ SBOM generated successfully
✅ No GPL licenses detected
✅ No high/critical vulnerabilities found

@github-actions

Copy link
Copy Markdown

Policy Coverage Report

Tools Coverage: 100%
Events Coverage: 100%
Egress Labeled: 100%
Replay Determinism: 100%

All policy gates passed successfully!

@fraware
fraware merged commit 3fd4fa4 into main Jul 23, 2026
135 checks passed
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.

2 participants