Repository navigation
Releases
macOS Inspector 2.2.0 - Saved PDF review and history coverage
macOS Inspector 2.2.0
Released October 25, 2026. This release completes saved PDF review in the document workspace and makes history-loading limits visible. Existing Mac audit, browser extension review, and static PDF inspection remain available.
Getting started
- Download
macos-inspector-2.2.0-macos.zipandSHA256SUMSbelow. - Verify the archive's SHA-256 against
SHA256SUMS, then extract it. - Double-click macOS Inspector.command and keep the launcher window open.
- Open PDF Inspector, expand Recent PDF inspections, and choose Review saved inspection beside a supported report.
Requires macOS and Python 3.10 or newer. Core operation needs no account, administrator privileges, mandatory Full Disk Access, or external service. Start the new package with your existing report output directory to review earlier supported reports. The launcher and installation guide explain output selection and first-launch behavior. The package is not Apple-signed or notarized; do not disable Gatekeeper globally.
Review saved PDF evidence without the original document
Review saved inspection loads the saved JSON evidence into the same PDF workspace used for a new inspection. The notice identifies the recorded analysis date and Inspector version. Actions, scripts, destinations, attachment references, metadata, and analysis limits remain available, with links to the original exports when present. Keyboard focus moves to the saved-evidence notice, and long fields wrap on narrow screens.
No PDF is uploaded, reopened, rendered, or reanalyzed. Loading does not authenticate the report, establish the document's current security status, or apply current parser rules to an older result. For example, an analysis from 2.0.0 retains that version and its interpretation limits; review of the report does not silently convert it into a 2.2.0 analysis. Inspect the original again when current checks are needed.
Use Clear to remove displayed evidence without deleting a saved report. Choosing another report, selecting a new file, or clearing the workspace invalidates pending loads and sharing previews. Late responses cannot restore a previously selected result, and mismatched response identities are rejected.
Prepare a sharing copy from a saved inspection
After loading saved evidence, press Preview sharing copy. Expand All fields in this sharing copy, review the complete reduced summary, then download HTML or JSON. The original document is not required.
The existing field allowlist omits detailed document evidence. Include document SHA-256 remains off by default and requires explicit selection. Original reports remain unchanged; downloads are generated in the browser, with no sharing copy saved on the server. A reduced copy is not anonymization, an authenticated original, a fresh analysis, or a safety verdict. Counts and optional fingerprints can still identify the document. The sharing summary's tool version identifies the software generating that copy, not necessarily the version that inspected the PDF.
For example, you can reopen last week's recorded inspection and prepare a reduced handoff without uploading the PDF again. Keep the detailed original report for investigation and read its original analysis date and limitations before drawing conclusions.
Understand history coverage
The history panel now separates loading coverage from PDF findings:
- Usable records remain visible when other reports cannot be read or validated.
- Selected but unavailable reports, rejected PDF-named entries, and older candidates not loaded have separate counts.
- Incomplete or interrupted directory enumeration is explicit.
- History selection details explains the selection method and limits.
- Failed refreshes remove stale rows and coverage details. Missing or inconsistent metadata from older servers is labeled unavailable rather than interpreted as zero.
Enumeration examines at most 5,000 directory entries, including unrelated files, and selects at most 25 canonical regular report candidates. Ordering uses filesystem modification time among examined candidates, not analysis date. Copying or touching a report can change its position. An invalid recent candidate still occupies a selection slot; older files are not read to fill the gap. Incomplete enumeration cannot establish which reports are newest in the entire directory.
For example, a list can show one usable report while explaining that another selected report could not be loaded. That is a partial history view, not evidence that the omitted document is safe or malicious. An empty list does not prove that no saved reports exist. No report is deleted, repaired, or reanalyzed.
Read boundaries and privacy
Saved review requires authenticated local API access. Report-download cookies alone cannot authorize the review API. Identifiers must match the canonical inspection filename shape, and JSON reads are limited to 2 MiB. Unsupported schemas, malformed display fields, invalid Unicode, oversized files, and symlinked reports are rejected. Nonregular files, including FIFOs, are rejected before reading saved-review or sharing evidence.
History diagnostics use counts and fixed explanations, without copying rejected filenames, paths, or raw exception text. Supported history rows and original reports still contain recorded filenames, fingerprints, and potentially sensitive evidence. No external destination is contacted, no original PDF is retained, and no host scan or process action starts when reviewing PDF history.
These checks are not report authentication, an atomic filesystem snapshot, or protection against a compromised same-user or administrative account. The PDF parser's existing limits remain: no rendering, code execution, decryption, destination contact, or complete final-revision reachability analysis. Recorded indicators do not establish execution or compromise; missing indicators do not establish safety.
Validation
All 169 Python tests, 94 browser tests across Chromium and WebKit, and 26 synthetic trust/response scenarios passed locally. Coverage includes authenticated read-only retrieval, original metadata retention, real saved-report review after page reload, reduced sharing previews without another upload, delayed and mismatched responses, mixed valid/invalid history, enumeration and selection limits, interrupted iteration, FIFO rejection before reading, stale-refresh cleanup, four display widths, keyboard focus, targeted accessibility, and extracted portable-launcher startup and shutdown.
The portable archive was built twice with identical bytes and checked for ZIP integrity. Syntax, documentation links, Git integrity, and release privacy checks were run. GitHub CI repeats Linux/macOS tests, Chromium/WebKit tests, and packaging before publication. Passing regression fixtures do not establish comprehensive browser compatibility, real-world malware detection accuracy, or resistance to every hostile PDF.
See the PDF Inspector guide, validation guide, and changelog for workflow details and limits. The published 2.1.0 tag and archive are unchanged.
Portable artifact
Filename: macos-inspector-2.2.0-macos.zip
SHA-256: 7cb3449bed37ad364eaacfc895e0da41b6cc397c6ea5b555807e63ce1f7c8acf
Verify the ZIP against the matching SHA256SUMS before launching. The package is not Apple-signed or notarized.
macOS Inspector 2.1.0 - Extension permissions and PDF sharing copies
macOS Inspector 2.1.0
Released October 9, 2026. This release explains browser extension permission declarations and adds previewed PDF sharing copies. Mac audit and PDF Inspector remain separate local workflows.
Getting started
- Download
macos-inspector-2.1.0-macos.zipandSHA256SUMSbelow. - Compare the ZIP's SHA-256 with
SHA256SUMS, then extract it. - Double-click macOS Inspector.command and keep the launcher window open.
- Choose Privacy and browsers for extension evidence, or PDF Inspector for document inspection and sharing copies.
Requires macOS and Python 3.10 or newer. Core operation needs no account, administrator privileges, or mandatory Full Disk Access. Protected browser sources can be unavailable, and collection limits remain visible. The package is not Apple-signed or notarized. Follow the installation guide if macOS blocks its first launch; do not disable Gatekeeper globally.
Browser extension permissions
After a browser audit, open Browser extension permissions in Simple or Analyst view. Each observed extension shows its recorded identity and version, required versus optional capabilities, host patterns, content-script restrictions, and evidence limits. Full HTML reports and investigation summaries include these explanations; original structured evidence remains available.
- Chromium-family browsers: inspect bounded Manifest V2/V3 declarations from the highest numeric version directory found in the local selection. This is an observed version, not confirmation that the browser runs it.
- Firefox: explain recorded addon permission metadata and enabled state when present. These records do not establish current runtime grants.
- Safari: retain limited legacy inventory and explicitly disclose that Safari app/WebExtension permissions were not assessed.
For example, an extension requesting all websites and cookie access deserves a review of its purpose and actual browser settings. Those requests alone do not prove access to every cookie, malicious behavior, or compromise. Optional permissions remain optional declarations, not confirmed grants. Native messaging does not prove that a host exists or has run. No extension is disabled or removed.
Unreadable, malformed, unsupported, symlinked, oversized, or truncated sources remain limited. Input sizes, directory enumeration, permission lists, normalized records, and per-profile output are bounded. Requested permissions are not a software reputation database, and no browser store or online service is contacted. Review reports before sharing: extension IDs, profile names, local paths, and host patterns can be sensitive.
See Browser extension permissions for the workflow, supported evidence, bounds, and interpretation examples.
PDF sharing copies
After a new PDF inspection, use Prepare a sharing copy and press Preview sharing copy. Review the preview and expand All fields in this sharing copy, then download the reduced HTML or JSON. Both exports contain the same summary fields and use generic filenames.
An explicit field allowlist retains the recorded assessment, structural-name counts, action-type counts, evidence-record counts, and scope caveats. It excludes document and attachment names, metadata, timestamps, inspection identifiers, destinations, file paths, JavaScript contents and hashes, raw context, and diagnostics. Include document SHA-256 is off by default and must be selected explicitly.
For example, you can share a summary that reports a JavaScript action without sending its script text, destination, or original filename. The reduced summary intentionally cannot replace the detailed evidence needed to investigate that action. Counts and optional fingerprints can still identify a document: this is not anonymization, an authenticated original, or a safety verdict.
Original reports and PDFs remain unchanged. Preview preparation uses the authenticated local API; downloads are generated in the browser, and no sharing copy is saved on the server. Changing the document or hash option invalidates the preview, including late responses. Historical reports retain their original private export links; inspect the document again to use the current-result sharing workflow.
See the PDF sharing guide.
Interpretation and request handling
OpenAction entries outside a recovered catalog no longer claim a document-open trigger. Their evidence remains visible with an explicit interpretation limitation. Ordinary catalog actions and their chains remain supported. The parser still does not validate final cross-reference reachability; recovered older revisions can contribute superseded evidence.
JSON APIs now require one bounded Content-Length, reject transfer-encoding conflicts and incomplete bodies, and limit body read time. Sharing previews reject missing, oversized, invalid, or symlinked source reports and require the same authenticated local session as other private APIs. Report-access cookies alone cannot authorize them.
Validation and remaining limits
All 162 Python tests, 88 browser tests across Chromium and WebKit, and 26 synthetic trust/response scenarios passed locally. Coverage includes real authenticated PDF previews and downloads, stale-response handling, hash opt-in, malformed and limited sources, escaped output, both audit views, four display widths, targeted accessibility checks, and extracted portable-launcher startup and shutdown. The portable archive was rebuilt deterministically and checked for integrity. GitHub CI repeats Linux/macOS unit checks, both browser engines, and packaging before publication.
These checks do not establish real-world malware-detection accuracy, comprehensive browser compatibility, effective extension grants, PDF anonymization, or resistance to every hostile document. The PDF parser does not execute scripts, render pages, decrypt content, contact destinations, or reconstruct every reader's final object graph. It is static triage, not an antivirus engine, sanitizer, or complete operating-system sandbox. Same-user or administrative compromise remains outside the local-session protection boundary. Review the validation guide and feature guides before drawing security conclusions.
The published 2.0.0 tag and artifact are unchanged. This release adds a normal commit on the existing history. It also includes the previously merged dead-code cleanup contributed by Cristian Chiriac in PR #1.
Portable artifact
Filename: macos-inspector-2.1.0-macos.zip
SHA-256: 407374231c7783d5e7cba8d34f7590ee03a07f39627bf175f50c4fa1e916a987
Verify the archive against the matching SHA256SUMS before launching. The package is not Apple-signed or notarized.
macOS Inspector 2.0.0 - PDF Inspector and private local sessions
macOS Inspector 2.0.0
Released October 9, 2026. This release adds local PDF triage and requires authorization for the dashboard API and HTTP report access.
Getting started
- Download
macos-inspector-2.0.0-macos.zipandSHA256SUMSfrom the assets below. - Verify the ZIP's SHA-256 against
SHA256SUMS, then extract it. - Double-click macOS Inspector.command. The launcher opens an authorized local browser session. Keep its window open while using the application.
- Use Mac audit for application, process, and system checks, or PDF Inspector for document analysis. Do not open the source HTML files directly.
Requires macOS and Python 3.10 or newer. No account, mandatory Full Disk Access, or administrator privileges are needed for core operation. Some collectors require additional permissions, which remain explicit. The package is not Apple-signed or notarized; see the installation guide before the first launch.
PDF Inspector
Inspect a PDF before opening it. From the audit page, choose PDF Inspector, select a local document, and press Inspect PDF. Selection alone does not start analysis. PDF analysis and its history have their own HTML page and module script, separate from Mac audit controls, process actions, and scan history. Opening the PDF page does not load audit code, Mac inventory, or audit history. Back to Mac audit returns to all existing system-investigation features; the browser session works across both pages in the same tab.
The redesigned document page includes clear selection, inspection, and review steps, grouped evidence, HTML/JSON export links, and labeled mobile evidence rows. Saved PDF reports are available after returning from the audit page; transient file selections are not guaranteed to survive navigation or refresh. The browser asks before leaving an in-progress analysis. Browser navigation and direct page links are supported. Small transitions and a busy indicator respect reduced-motion preferences; no simulated progress percentage is shown.
The report records PDF headers and recovered objects, parsed structural names, JavaScript strings and supported streams, declared open and interaction actions, action chains, destinations, attachment references, and document metadata. Supported compressed object streams and escaped names are inspected. Script excerpts and records are bounded. The uploaded bytes receive a SHA-256 identity; HTML and JSON reports stay local and recent inspections have their own history.
No PDF viewer is opened, JavaScript is not executed, attachments are not launched, and destinations are not contacted. A URI link is not an observed web request. Document-open actions, document-level script registration, additional events, and user interaction are shown separately. No result labels a document safe or malicious. Encryption, unsupported filters, malformed objects, revisions, unresolved references, and analysis bounds restrict interpretation explicitly.
The original PDF is not retained by Inspector. Reports can contain sensitive document data. The built-in parser needs no extra package or online service. It is static triage, not a conformance validator, antivirus engine, sanitizer, or complete OS sandbox.
Reading a result
- An ordinary invoice link appears as a declared destination. Inspector does not visit it or claim that a request occurred.
- A JavaScript open action appears with its declared trigger, object context, script excerpt, and interpretation limits. This is a reason to review the source, not proof of malware or reader execution.
- Encrypted content or unsupported filters restrict analysis. An incomplete result cannot establish that JavaScript, actions, or attachments are absent.
The PDF Inspector guide covers the workflow, practical examples, supported structures, privacy behavior, and limitations.
Consolidation of the 1.4.4 foundation
Failed YARA commands no longer produce Pass merely because stderr is empty. Collection errors now contribute to recorded coverage limits. Local server binding no longer performs reverse DNS, preventing slow hostname resolution from delaying dashboard or portable-launcher startup. Existing application trust, process follow-up, rule provenance, report-selection, and guarded process-response checks remain in place. The published 1.4.4 tag and history are not rewritten.
Private local sessions
The launcher authorizes the browser using an owner-only per-launch link without a user account or password. The launch fragment is exchanged and removed from the address. API credentials are origin-scoped and are never attached to cross-origin requests. Report access uses an independent HttpOnly, SameSite=Strict cookie with a private path. Report cookies cannot authorize API calls or process actions. Restarting the server invalidates prior credentials.
This is a breaking change for unauthenticated API integrations and old direct HTTP report links. API callers must exchange the launch credential using POST /api/session and send the returned API token as a Bearer header. Reopen the launcher to authorize another browser session. Reports saved on disk remain readable locally.
Verification and limits
All 143 Python tests passed using an existing runtime with optional cryptographic support. The default local runtime skips only the two optional cryptography checks. All 82 browser tests across Chromium and WebKit and all 26 synthetic trust/response scenarios passed. This includes real authenticated PDF uploads and report downloads in both browser engines, an extracted portable-launcher/worker check with shutdown cleanup, static PDF features, malformed/limited input, private sessions, DNS-independent loopback startup, worker execution, HTML escaping, responsive layouts, separate-page isolation/navigation, interrupted-navigation confirmation, reduced motion, and targeted accessibility. A read-only Security and Live Triage collection on the local Mac completed without collection errors and its evidence manifest verified. These checks do not establish real-world malware-detection accuracy or eliminate every parser risk. CI is configured to repeat Linux/macOS unit checks, Chromium/WebKit checks, and deterministic packaging before publication.
PDF uploads are capped at 25 MiB, one concurrent analysis, bounded objects/tokens/nesting/decoded bytes/evidence, a ten-second upload read timeout, a twenty-second wall-clock worker timeout, and a ten-second worker CPU limit. Linux adds a 512 MiB address-space cap; macOS relies on algorithmic and time bounds rather than that hard cap. Same-user or administrative compromise remains outside the local-session protection boundary.
Portable artifact
Filename: macos-inspector-2.0.0-macos.zip
SHA-256: 451411e29c84014a2efd628a149a0101a283338988de073f0c59caa4c6a6d4a4
Verify the archive against the matching SHA256SUMS before launching. The package is not Apple-signed or notarized.
macOS Inspector 1.4.4
macOS Inspector 1.4.4
This release makes process follow-up more precise, shows the local rule context behind a result, and keeps delayed dashboard responses from mixing different scans.
Process follow-up
After a recorded process action, use Process action history > Recheck process outcome. The follow-up is a new read-only Live Triage snapshot, not another termination request.
The result distinguishes the original identity still present, a reused PID, a zombie, the original PID no longer observed, and insufficient evidence. It now also shows when the same executable path and numeric owner appear under another PID. That observation may describe an existing instance or a restart; it proves neither. A matching path does not establish identical file contents or malicious activity.
Follow-ups require matching recorded host context and valid, timezone-aware action, scan-window, and process-evidence timestamps. Prior or overlapping scans, duplicate PIDs, incomplete inventories, and missing historical context remain unverifiable. New actions record the local hostname. Older action records remain visible but cannot supply missing identity or host evidence.
Normal Live Triage process results use Observed, which the follow-up now accepts correctly. Process inventory completeness is recorded explicitly. Unparsed rows, duplicate PIDs, missing numeric owners, command failures, and collection bounds prevent absence inferences without discarding the processes that were observed.
Local rules and provenance
When IOC or YARA was selected, About this scan > Local rules and provenance shows the recorded rule snapshot separately from result availability. The same context is included in the standalone HTML report and investigation summary.
Expand Recorded rule details and provenance for the context SHA-256, fingerprinted file count, and bounded IOC pack metadata. Pack versions, update labels, and source hostnames are declared by the pack, not independently authenticated. Source URL credentials, query strings, fragments, and full source paths are not copied into the derived summary. Opening the panel does not contact a source.
A stable fingerprint does not mean the rules are reliable or that checks ran. Disabled YARA can have a stable fingerprint and still say Not run. Filtering can omit metadata or results; missing context remains explicit rather than implying a clean assessment.
Consistent report selection
Report evidence, decision support, and export links load as one selected snapshot. Delayed responses from an earlier selection cannot overwrite a newer scan. Decision support with a different scan ID is rejected while the recorded report remains available.
Process response is blocked while a report is loading. Investigation saves capture the originating scan and do not replace another scan's guidance after navigation. Candidates that are no longer displayed are removed from the dashboard's process-action map. Server-side process authorization and identity checks remain independent.
Verification
Local verification passed 121 Python tests, 68 browser tests across Chromium and WebKit, and all 26 synthetic scenarios. Browser checks include delayed JSON and decision-support responses, report identity mismatches, pending-load process controls, investigation saves during navigation, narrow layouts, escaped provenance, and targeted accessibility checks.
The portable package was rebuilt deterministically, checked after extraction, and tested with local read-only collection. Git integrity and privacy checks passed. GitHub CI repeats Linux/macOS tests on Python 3.10 and 3.13, both browser engines, and portable packaging. Publication follows a successful CI run and a byte-for-byte comparison of its artifact with the local build.
Boundaries
Collectors remain read-only. Process signals still require explicit confirmation, a recent authorized snapshot, current-user ownership, and live identity revalidation. No automatic response, reputation database, telemetry, new online provider, or evidence upload was added.
Hostnames and owner-controlled logs are not authenticated identities or tamper-proof evidence. Process collection and rule capture are not atomic. An absent PID does not establish containment or resolve an investigation. Online intelligence remains opt-in.
The dashboard assumes a trusted local session. Browser-origin and loopback checks are not per-client authentication; another local program or account can call the API. Close the dashboard when finished and do not run it as a shared service on a host with untrusted local users. This release clarifies that boundary; it does not add client authentication.
The action log retains one previous file and shows up to 50 recent actions. Source findings remain unchanged. The 26 scenarios are regression fixtures, not an independent detection benchmark. WebKit automation does not establish coverage for every Safari version or screen reader.
Python 3.10 or newer is still required. The portable release is not Apple-signed or notarized. No runtime installation, privilege elevation, or change to macOS security settings is automatic.
Install and artifact
Download the ZIP and SHA256SUMS from this release, compare the complete SHA-256, extract the archive, and double-click macOS Inspector.command. If a compatible runtime is missing, use the included docs/START_HERE.html.
Filename: macos-inspector-1.4.4-macos.zip
SHA-256: a9f8b6734c318bab25d40b95b09ad026c790324d213e6927b59a5eb2fc435618
See the README, Usage guide, Validation guide, and Changelog.
macOS Inspector 1.4.3
macOS Inspector 1.4.3
This release makes scan comparisons easier to interpret and adds a traceable follow-up after an explicit process action. Both dashboard views now show when evidence was collected, what was checked, and which differences the available evidence can actually support.
What changed
Recorded scan context and repeat checks
- About this scan shows start and completion times, report age, selected sections, exact application target, tool version, and finding filter.
- Run these checks again preserves the recorded sections and application target, uses current settings and case fields, and still asks before an online check. Unsupported scope and duplicate requests are blocked.
- Changes since last comparable scan is visible in Simple and Analyst views. It shows the actual earlier scan ID and completion time, with View earlier scan, expandable highlights, and all comparison counts.
- Baseline selection uses timezone-aware completion times and matching recorded hosts, collection sections, application targets, and report identities. A matching hostname is not device authentication.
More careful comparison and inventory handling
- Unreadable application directories are recorded as discovery gaps while successfully discovered applications remain in the report. They no longer disappear silently from the inventory.
- An unknown result or collection error cannot round completion up to 100%.
- Filtered, incomplete, or missing collection evidence makes affected additions/disappearances Not comparable instead of zero. Differences between records actually present in both scans remain available.
- Local IOC/YARA files and relevant nonsecret configuration receive before/after content fingerprints. Missing, incomplete, unstable, or changed fingerprints prevent unsupported finding-presence and numeric-index comparisons.
- Manual comparisons use Baseline only, Comparison only, and Changed record, not resolved-incident wording. They preserve raw differences and show N/A for unsupported index deltas. Standalone comparison reports and investigation summaries include the same interpretation limits and snapshot times.
- Security-control regressions remain visible even when an unrelated comparison section is incomplete.
Process action history and outcome rechecks
- Process action history shows recent locally recorded signals in both views, with access to the preserved source report.
- Recheck process outcome collects a new, unfiltered Live Triage snapshot linked to that action. It is read-only and sends no signal.
- Follow-up distinguishes the same identity still observed, PID reuse, zombie state, the original PID not observed, and insufficient evidence. No result marks an investigation resolved automatically.
- New action records and bounded process snapshots include numeric owner and start identity. Older logs remain visible, but missing identity data cannot establish an identity-aware outcome.
Validation and setup
- Detection regression validation evaluates 18 bundled synthetic trust and response scenarios from the HTML page and downloads a JSON result. These fixture counts are not real-world malware detection accuracy.
- Dashboard tests now run in Chromium and Playwright WebKit, with targeted automated accessibility checks, keyboard focus, narrow layouts, escaped evidence, linked follow-up, and download checks.
- The double-click launcher tries compatible Python installations instead of stopping at an incompatible first candidate. If none is found, it opens offline setup help. No runtime, dependency, or software is installed automatically.
- Reopening an existing dashboard no longer removes the first launcher's PID tracking file.
How to use the new workflow
- Open the dashboard with macOS Inspector.command and run Check this Mac.
- Read About this scan before interpreting its assessment or comparing it with an earlier report.
- Use View earlier scan to inspect the actual automatic baseline. Read any Not comparable reason before treating a missing application or finding as a change.
- Before containment, preserve relevant evidence and validate the process. Confirm Terminate or Force kill only when justified.
- In Process action history, select Recheck process outcome and open the linked later snapshot. A missing PID does not prove that related activity stopped or that a security concern was fixed.
Verification
Local verification passed 113 Python tests, 50 browser tests across Chromium and WebKit, and all 18 synthetic scenarios. Syntax, compilation, documentation links, deterministic rebuilds, extracted-package startup, Git integrity, and privacy checks also passed. GitHub CI repeats Linux/macOS tests on Python 3.10 and 3.13, both browser engines, and portable packaging before publication. The CI artifact is checked against the local package byte for byte.
Boundaries
Collectors remain read-only. Process response still requires explicit confirmation, a recent authorized snapshot, current-user ownership, and live identity revalidation. No reputation database, automatic remediation, telemetry, or evidence upload was added. Online intelligence remains opt-in.
The action log is private but not tamper-proof and retains only one previous file; the dashboard shows up to 50 recent actions. Rule capture is not an atomic filesystem snapshot. Playwright WebKit coverage is not a claim that every shipping Safari version or screen reader was tested. Synthetic fixtures are regression checks, not an independent detection benchmark.
The portable release still requires Python 3.10 or newer and is not Apple-signed or notarized. Protected evidence can remain unavailable without broader permissions; do not disable macOS protections just to make checks green.
Install and artifact
Download the ZIP and SHA256SUMS from this release, compare the full SHA-256 before extraction, then double-click the launcher. If Python is missing, open docs/START_HERE.html for setup instructions.
Filename: macos-inspector-1.4.3-macos.zip
SHA-256: d4fcff9aab678fd6d7687c300a3c0c501db6ee4fc149c23a8f349ccc69eebf21
See the README, Usage guide, Validation guide, and Changelog for complete instructions and changes.
macOS Inspector 1.4.2
macOS Inspector 1.4.2
This release tightens two decisions that matter during an investigation: whether an application trust check actually finished, and whether a process shown in a report is still the process about to receive a signal.
Changes
- Application Trust now records an incomplete code-signature or Gatekeeper check as unknown when macOS returns an internal trust-service error. A separate file-permission or integrity concern remains visible even if a trust check did not finish. The dashboard names the missing check and lowers evidence confidence instead of implying that the app passed or failed it.
- A validly signed Apple system component that Gatekeeper says is not an independently assessable app is shown as Gatekeeper not applicable. That result is not a rejection and does not establish that the component is safe.
- Process response requires a Live Triage snapshot no older than 15 minutes. Immediately before sending a signal, macOS Inspector compares the current PID, owner, parent, executable path, and process start with the report. Old reports and snapshots without a start identity cannot authorize a process action.
- An isolated low-priority process signal, such as a path under a hidden user directory, stays in the technical evidence as context. It no longer creates a process-response finding on its own. Corroborated medium- and high-priority candidates still appear for review.
Using this release
If an application card says a trust check did not finish, use Recheck this app before deciding whether to keep using the application. Review any separate permission or integrity warning on the same card; a missing trust result does not erase it.
If process-response controls say the snapshot is stale, run Live Triage again. Confirm the process is unexpected and preserve relevant evidence before terminating it. A process marked zombie has already exited; signaling it will not remove it.
Boundaries
The collectors remain read-only. Process termination is a separate, explicitly confirmed action; the identity check narrows PID-reuse risk but cannot eliminate the race between verification and signaling. No application reputation database was added. Online intelligence remains opt-in and does not turn an indicator match or signature result into a malware verdict.
The portable package is not Apple-signed or notarized. Verify its checksum before opening the launcher. This release does not require administrator privileges or Full Disk Access for normal use, though protected evidence may remain unavailable.
Install
Download macos-inspector-1.4.2-macos.zip and SHA256SUMS from this release. Run shasum -a 256 macos-inspector-1.4.2-macos.zip and compare the full digest with SHA256SUMS. Extract the ZIP, then double-click macOS Inspector.command.
Artifact
Filename: macos-inspector-1.4.2-macos.zip
SHA-256: 40a7ed7fc2d36f99c2c725456a11dd5db9bbd9209f88445cc039ca7558fa0b3f
See the README, Installation guide, Usage guide, and Changelog for setup and interpretation.
macOS Inspector 1.4.1
macOS Inspector 1.4.1
Version 1.4.1 makes incomplete and out-of-scope checks harder to mistake for clean results. It also improves how older Application Trust reports are presented and makes the portable build reproducible.
What changed
Not Applicableis shown as Not assessed in guided review. Disabled YARA scanning, an absent IOC pack, and a missing supported browser profile now lead to a concrete setup or scope check.UnknownandNot Applicablefindings remain visible when a minimum-severity filter is used. A failed collector or one that returns no findings is included in the collection-completion calculation.- A scan with no assessed findings shows
N/Afor its rule outcome index in history and human-readable reports. JSON retains the raw score and records assessed-finding counts. Comparisons showN/Afor an index delta that cannot be interpreted, while still showing finding-level changes. - A category made entirely of
Not Applicableresults no longer raises the overall rule outcome index. AnUnknowncategory still reduces the raw index because evidence for a selected check was unavailable. - The dashboard calls its percentage Collection completion. It describes whether selected checks returned statuses, not how many files, applications, or activities were assessed. Exported category tables call this Status availability.
- Older Application Trust findings without recorded check-completion evidence get a recheck cue and lower evidence confidence. The original report and finding status are not rewritten.
- IOC and YARA matches remain single pieces of evidence. They do not establish malware or unwanted software by themselves.
- Chromium smoke tests cover the guided and analyst views, incomplete scans, history, and narrow browser layouts.
- The portable ZIP uses fixed file metadata and ordering so identical source content produces identical bytes.
How to use the new scope information
Start with the current assessment and the recommended next steps. If a check says Not assessed, open it to learn what target or optional configuration was missing. If it says Unknown, inspect collection readiness and rerun the same scope when access is available. A 100% collection-completion value can coexist with Not assessed results; it is not a safety verdict or a measure of target coverage.
For an older Application Trust warning, use the focused recheck for that application before deciding whether its earlier negative result reflects a current trust failure.
Privacy and limitations
Local scans remain offline unless an online intelligence feature is explicitly enabled. This release does not add an application reputation database. Code signatures, Gatekeeper results, and indicator matches require context; none proves that an application is safe or malicious. Protected macOS evidence may remain unavailable without the relevant user permissions.
The portable release is not Apple-signed or notarized. Verify the downloaded ZIP before opening the launcher. Process termination remains a separate, explicitly confirmed action; scanning and rechecking do not terminate processes or remove software.
Install
- Download
macos-inspector-1.4.1-macos.zipandSHA256SUMSfrom this release. - Calculate
shasum -a 256 macos-inspector-1.4.1-macos.zipand compare the full digest withSHA256SUMS. - Extract the ZIP and double-click macOS Inspector.command.
Artifact
Filename: macos-inspector-1.4.1-macos.zip
SHA-256: 887b03f7fac6a6fccbc2f3d9511d085fa0d632b0c477c59a6ad57faeb83929ba
Read the README, Installation guide, Usage guide, and Changelog for setup, interpretation, and known limitations.
macOS Inspector 1.4.0
macOS Inspector 1.4.0
macOS Inspector 1.4.0 makes the dashboard easier to use without hiding the evidence an experienced investigator needs. It also fixes an Application Trust timeout that could make a large, valid application appear to have an invalid signature.
What is new
- Simple view keeps the recommended checks, focused application scan, current assessment, application review, reports, and scan history in one guided workflow.
- Analyst view exposes the complete collector, evidence, export, comparison, and case controls.
- The dashboard now gives one of four plain-language assessment states: no immediate warning identified, needs review, action recommended, or scan incomplete.
- The response plan shows no more than three priority items. Each item explains the observation, why it matters, what it does not prove, how to verify it safely, and the risk of taking action.
- Review priority, collection coverage, and evidence confidence are shown separately. The existing numeric rule outcome index remains available for compatibility and is not presented as a probability that the Mac is safe or compromised.
- Deep code-signature checks now have their own bounded timeout. If a check cannot finish, the result is recorded as incomplete instead of being reported as an invalid signature.
- Mac App Store acceptance is kept separate from a notarization result that Gatekeeper did not report.
The Xcode verification fix
Large application bundles can take longer to verify than smaller apps. Earlier releases used a generic 15-second command limit and interpreted an unfinished deep signature check as a failed check. On the development Mac, Xcode completed Apple's deep signature verification successfully in about 14.3 seconds and Gatekeeper accepted it as a Mac App Store application.
Version 1.4.0 allows up to 60 seconds for deep signature verification and up to 30 seconds for the Gatekeeper assessment. It records whether each command completed, timed out, and returned a result. Old reports with the timeout pattern are displayed as Signature check not completed, not Signature invalid. Run a new focused Application Trust scan to obtain a current result.
How to use it
- Open the dashboard and leave Simple selected if you want a guided check.
- Choose what you noticed or use Quick triage.
- For one app, select it under Check one application and run the focused trust check.
- Read Current assessment first, then work through the listed verification steps.
- Switch to Analyst when you need raw findings, evidence, comparisons, case metadata, or additional export controls.
A warning is a reason to investigate. It is not proof that an application is malicious or that the Mac is compromised.
Online intelligence and privacy
This release does not add an application reputation database and does not claim to know every good or bad application. Local results are based on identity, signature, Gatekeeper, provenance, integrity, behavior, change history, and the evidence collected on this Mac.
Online intelligence remains optional. Supported free sources are disabled unless the user selects or configures them. Hash reputation checks send only the selected file digest to the chosen provider, never the application file. Local scans continue to work without internet access.
Install
- Download
macos-inspector-1.4.0-macos.zipandSHA256SUMSfrom this release. - Calculate
shasum -a 256 macos-inspector-1.4.0-macos.zipand compare the complete result withSHA256SUMS. - Extract the archive and double-click macOS Inspector.command.
The portable release is not Apple-signed or notarized. Verify the checksum before opening it. If macOS blocks the first launch, use Finder's Control-click, then Open. Do not disable Gatekeeper globally.
Artifact
Filename: macos-inspector-1.4.0-macos.zip
SHA-256: 1bfe1c5aed9dfc223288e076c04b9e3c4dd4206b668616ad2807a5d6d65ea561
Known limitations
- An incomplete trust check cannot establish whether a signature is valid or invalid. Recheck the application and inspect the command status in Analyst view.
- Gatekeeper acceptance and code-signature validity do not prove that an application is safe.
- Mac App Store acceptance does not provide a separate notarization result in this workflow.
- Some protected evidence requires Full Disk Access. Missing access reduces coverage and is reported as such.
- Process containment remains a deliberate, confirmed action. macOS Inspector does not remove applications or change system security settings.
- The portable release is not Apple-signed or notarized.
Read the README, Installation guide, Usage and interpretation guide, Threat model, and Changelog before using the tool in an investigation.
macOS Inspector 1.3.3
macOS Inspector 1.3.3
macOS Inspector 1.3.3 makes comparisons more reliable when the baseline and current scan were created by different versions of the tool.
What is new
- The dashboard shows the macOS Inspector version recorded by each compared report when those versions differ.
- A comparison note explains that collector coverage and detection rules can change between releases.
- Evidence newly evaluated by a later version is no longer automatically counted as proof that the application changed.
- New concerns produced by expanded checks remain visible and keep their security priority, but their explanation states that the underlying application change has not been established.
- Coverage updates are counted separately from changed applications.
- The Application Review queue includes a New evidence coverage filter when relevant.
- Standalone investigation summaries include the same version and coverage context.
Why this matters
An older report may contain fewer integrity checks than a newer report. For example, a later release may start evaluating permissions that were already present during the earlier scan. The new comparison logic distinguishes that situation from a known field changing between scans.
Current trust concerns still require validation. The distinction only prevents the dashboard from claiming that the application itself changed when the available evidence does not establish that conclusion.
How to use it
Open a completed scan with an earlier comparable baseline. If the reports were created by different versions, read the Comparison note before interpreting the change counts.
Use Changed since last scan for applications with established field changes. Use New evidence coverage for results introduced by expanded collection or detection. Open the current result to review the complete trust and integrity evidence.
Safety and privacy
Comparison processing remains local. This release does not add file uploads, automatic removal, quarantine, or background network requests.
Online OSINT remains opt-in. Process containment still requires explicit confirmation in the local dashboard.
Install
- Download
macos-inspector-1.3.3-macos.zipandSHA256SUMSfrom this release. - Run
shasum -a 256 macos-inspector-1.3.3-macos.zipand compare the complete result withSHA256SUMS. - Extract the archive and double-click macOS Inspector.command.
The portable release is not Apple-signed or notarized. Verify the checksum before opening it. If macOS blocks the first launch, use Finder's Control-click, then Open. Do not disable Gatekeeper globally.
Artifact
Filename: macos-inspector-1.3.3-macos.zip
SHA-256: 8061c2a46c697668210650fba134e60efede2f3e42d00fcffadd0ecd39619972
Known limitations
- The distinction depends on the metadata and comparable field values stored in both reports.
- A coverage update can expose a real current problem even when it does not establish when that problem began.
- Reports without a recorded tool version cannot display the version comparison note.
- The portable release is not Apple-signed or notarized.
Read the README, Installation guide, Usage and interpretation guide, Threat model, and Changelog before using the tool in an investigation.
macOS Inspector 1.3.2
macOS Inspector 1.3.2
macOS Inspector 1.3.2 makes changes between Application Trust scans easier to understand and investigate.
What is new
- Application changes now explain whether the evidence is consistent with an expected update, a macOS update, changed executable contents, a new trust concern, a signing identity change, or a trust regression.
- Version, Team ID, signature, Gatekeeper, notarization, hardened runtime, signature type, executable hash, permissions, bundle path integrity, trust result, and severity are compared when relevant.
- High-priority trust regressions, signer replacements, bundle identity changes, and integrity problems appear before routine update context.
- A current failed, review, or unknown trust result cannot be hidden by a possible update label.
- Weakened executable or Info.plist permissions and executables that resolve outside their application bundle receive dedicated explanations and next steps.
- Applications that no longer appear in the current inventory are counted and shown as review items. The dashboard asks for confirmation instead of assuming that the application was removed or that its disappearance is malicious.
- The Application Review queue includes a Changed since last scan filter. The complete change set remains available even when the summary is limited to the highest-priority items.
- Standalone investigation summaries include the same change explanations and field comparisons.
How it works
Open a completed Application Trust scan that has an earlier scan with the same collection scope. The dashboard automatically selects the newest comparable scan and places the most important changes in What changed and how the evidence connects.
Use Changed since last scan to review every current application with changed evidence. Expand a change to compare the previous and current values, then follow the displayed validation step. Applications that are no longer present appear in the change summary because there is no current application card to open.
These labels prioritize investigation. A possible update label does not prove that an update is legitimate, and a changed or missing application is not proof of tampering.
Safety and privacy
This release does not add file uploads, automatic removal, quarantine, or background network requests. Comparison logic uses reports already stored in the local report directory.
Online OSINT remains opt-in. Process containment still requires an explicit confirmation in the local dashboard.
Install
- Download
macos-inspector-1.3.2-macos.zipandSHA256SUMSfrom this release. - Run
shasum -a 256 macos-inspector-1.3.2-macos.zipand compare the complete result withSHA256SUMS. - Extract the archive and double-click macOS Inspector.command.
The portable release is not Apple-signed or notarized. Verify the checksum before opening it. If macOS blocks the first launch, use Finder's Control-click, then Open. Do not disable Gatekeeper globally.
Artifact
Filename: macos-inspector-1.3.2-macos.zip
SHA-256: f7072237250b9be2f15cbe395469ee1033afd8f9c951e10e6a071d6a46f062e3
Known limitations
- A comparison depends on equivalent collection scope and the evidence available in both scans.
- Missing applications can reflect intentional removal, access differences, or inventory changes. Confirm the reason before drawing a security conclusion.
- A matching Team ID and version change support update context but do not authenticate the release by themselves.
- Signature, Gatekeeper, notarization, and filesystem observations describe the inspected state. They do not label an application as malware or safe.
- The portable release is not Apple-signed or notarized.
Read the README, Installation guide, Usage and interpretation guide, Threat model, and Changelog before using the tool in an investigation.