Commit Graph
15 Commits
Author SHA1 Message Date
Michael D'Angelo 8f18784d94 feat(endor): import findings (#1689)
Import complete Endor Finding JSON through the shared flow for local storage and Cloud publication. Prefer scoped, unmasked `endorctl api list --list-all` exports; self-contained SARIF remains a lossy compatibility input.
2026-10-22 19:12:58 +08:00
Michael D'Angelo 00bfe9d4ec feat(semgrep): import Code findings (#1690)
Import Semgrep Code/SAST CLI or CI JSON through the shared flow for local storage and Cloud publication, with SARIF compatibility. Platform API and Guardian/MCP response envelopes require a separate mapping.
2026-10-22 18:24:11 +08:00
Michael D'Angelo 0391e838c5 feat(snyk): import Open Source JSON findings (#1687)
Import Snyk Open Source dependency vulnerabilities from native CLI JSON or the official Codex plugin's JSON bundle, with SARIF compatibility. Uses the shared local and Cloud import flow.
2026-10-22 17:07:05 +08:00
Michael D'Angelo 04df406934 fix: preserve Daybreak settings across scans and remediation (#1740) 2026-10-20 09:12:27 +08:00
Michael D'Angelo 0740e97c1c feat(scan): add unified max-time investigation budgets (#1674) 2026-10-20 04:43:28 +08:00
Michael D'Angelo 40a8e7b52c docs: make security workflows discoverable to agents (#1698) 2026-10-20 04:39:28 +08:00
aiapienthusiastandMichael D'Angelo 07edb95dca feat: add Cheaper Inference inference provider (#1623)
Co-authored-by: Michael D'Angelo <mdangelo@openai.com>
2026-10-19 15:14:52 +08:00
Michael D'AngeloandCodex dae9d40746 feat(publish)!: share issue receipts with finding tracking (#1426)
* feat!: share finding issue receipts across publication and tracking

* fix(tracking): recover issue receipts after plugin restart

* fix(tracking): recover stored history before initial writes

* fix(tracking): preserve fallback for inaccessible primary stores

* fix(tracking): defer inaccessible receipt stores to SQLite

* refactor: derive publication membership from the sealed scan

* Preserve accepted receipt URLs and complete receipt diagnostics

* Index finding issue ownership checks without rewriting migrations

* test(deep-scan): isolate the Node capability bundle probe

* test(package): adapt Deep Scan fixture to direct engine

* fix(tracking): allow full-scan receipt work to complete

---------

Co-authored-by: Codex <codex@users.noreply.github.com>
2026-10-17 04:58:04 +09:00
Michael D'Angelo c380befb9e fix(scans)!: simplify scan outcomes (#1633)
Simplify scan outcomes across the CLI, SDK, and GitHub Action. Completed report-only scans return 0, an opt-in findings threshold returns 1, and execution or required-output failures return 2. Remove partial and unknown outcomes while retaining coverage diagnostics and saved results.

Require the current plugin for Deep Scan and canonical fields for new findings. Preserve readability of historical sealed reports. Update documentation and tests, and remove obsolete compatibility paths.
2026-10-14 11:32:39 -07:00
mldangelo-oai c30c00c9a9 fix(plugin): migrate Atlassian app and simplify finding workflows (#1039)
* fix(plugin): migrate Jira workflows to the current Atlassian app

* docs(plugin): make connector guidance concise and evergreen

* fix(plugin): simplify tracking with the current Atlassian app

* fix(plugin): remove the fixed finding batch limit

* refactor(plugin): simplify finding workflow prompts

* fix(plugin): preserve security policy in finding triage

* fix(plugin): stream helper arguments in read-only sandboxes

* fix(plugin): preserve helper cancellation and source checks

* fix(plugin): refresh reuse candidates after approval
2026-10-06 01:23:04 -04:00
mldangelo-oai c0f5a5b940 feat: support per-scan Cyber access programs (#1185)
Forward optional Cyber access selection through scans and fresh or resumed workers while retaining per-scan isolation.
2026-10-02 16:02:21 -07:00
mldangelo-oaiandmldangelo-oai 5bd200978c fix(bedrock): improve authentication and Blue/Red support (#1188)
* fix(bedrock): clarify authentication and support model-specific costs

* fix: preserve Bedrock command authentication diagnostics

* fix(bedrock): support Daybreak Red cost estimates

---------

Co-authored-by: mldangelo-oai <269034524+mldangelo-oai@users.noreply.github.com>
2026-10-02 12:38:32 -07:00
mluck-openai 794e046456 revert: restore GPT-5.6 Sol as the default model (#1181)
Revert PR #1078 while retaining Deep Scan worker model propagation assertions for the restored default.
2026-10-01 17:34:24 -04:00
mluck-openaiandmluck-openai b43d3dcd96 feat: default CLI and SDK to GPT-6 Sol xhigh (#1078)
* feat: default CLI and SDK to GPT-6 Sol xhigh
---------

Co-authored-by: mluck-openai <329245017+mluck-openai@users.noreply.github.com>
2026-09-28 14:40:21 -04:00
mldangelo-oaiandmldangelo-oai 408ce4465e feat: share scan settings across the CLI and SDK (#742)
* feat(cli): add YAML and JSON project configuration

* refactor(cli): simplify project configuration and docs

* fix(cli): repair configuration regressions and Windows tests

* feat(sdk): share project configuration with the CLI

Resolve project files and typed configuration into shared SDK scan settings.
Keep native Codex configuration separate from scan options and expose public
loadProjectConfig and resolveProjectConfig entry points.

Reuse prompt-file handling and severity comparison across SDK and CLI paths.
Preserve existing flags, file fields, defaults, output, and input protections.
Cover equivalent inputs, prompt-backed workflows, and installed package APIs.

* fix(sdk): allow reruns after blank scan prompts

Only record additional scan instructions as required when the prompt builder
actually includes them. Empty and whitespace-only SDK inputs should remain
replayable, matching omitted prompts and empty prompt files.

Exercise saved SDK recipes through the CLI rerun path while preserving the
requirement to resupply real additional instructions.

* refactor(config): use snake_case project keys

* fix(sdk): leave missing ambient deep config unchanged

* fix(sdk): compare normalized ambient config paths

* fix(config): preserve path aliases and repeated YAML profiles

* fix(config): address project configuration review feedback

Preserve workflow identity and ambient defaults, restore bounded YAML
expansion, and make required scan instructions replaceable on rerun.

Complete configuration support for bulk and component scans, add starter
generation and inspection, and centralize setting and provenance metadata.
Document the operator trust boundary and verify the installed Node package.

* fix(config): finish bulk inputs and portable starter checks

Use the actual scan context for prompt protections, reject unsupported bulk
scopes before starting work, and make nested starter schema hints relative.

Keep path fixtures native on Windows and compare macOS temporary aliases
canonically in the installed-package smoke check.

* fix(config): preserve bulk prompt repository checks

* fix(cli): preserve imported scan prompt errors

Keep the existing validation-option error when rerunning imported scans and identify the scan-instructions option when it is supplied. Cover both rejected inputs without launching an import.

---------

Co-authored-by: mldangelo-oai <269034524+mldangelo-oai@users.noreply.github.com>
2026-09-10 20:45:38 -07:00