Aegis
Local-first desktop monitoring for AI agents. Inspect processes, files, network activity and evidence; review selected actions through opt-in policy and MCP tools. Windows-first; macOS/Linux experimental.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 11 Nov 2026.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/antropos17/Aegis with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/antropos17/Aegis'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/antropos17--aegis)HTML
<a href="/listings/antropos17--aegis"><img src="/api/listings/antropos17--aegis/badge.svg" alt="findsafeskills safety rating" /></a>Install
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 367 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- <p align="center">
- <img src="docs/images/github-header.svg" width="960" alt="AEGIS. Your agents have access. You deserve visibility. A local security dashboard for AI coding agents.">
- </p>
- <a id="aegis--ai-agent-monitoring-for-windows"></a>
- # AEGIS — See what your AI agents are doing
- **Your agents have access. You deserve visibility.**
- AEGIS puts local AI-agent activity in one desktop dashboard. Follow detected
- Claude Code, Codex, Cursor and other agent processes, investigate sensitive-file
- observations and network connections, and review a project or skill before you
- run it. Open source. Windows first. No subscription for monitoring.
- <p align="center">
- <a href="https://github.com/antropos17/Aegis/releases"><strong>Download the Windows beta →</strong></a>
- ·
- <a href="https://antropos17.github.io/aegis-media/">Explore AEGIS</a>
- ·
- <a href="docs/screenshots/README.md">See the interface</a>
- </p>
- <p align="center">
- <a href="https://github.com/antropos17/Aegis/actions/workflows/ci.yml"><img src="https://github.com/antropos17/Aegis/actions/workflows/ci.yml/badge.svg?branch=master" alt="CI status"></a>
- <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-22d3ee" alt="MIT license"></a>
- <a href="#install"><img src="https://img.shields.io/badge/Windows-beta-f59e0b" alt="Windows beta"></a>
- <a href="SECURITY.md#privacy-architecture"><img src="https://img.shields.io/badge/monitoring-local-22d3ee" alt="Monitoring data stays local"></a>
- </p>
- [](docs/screenshots/README.md)
- *Interface preview with simulated data, captured 11 October 2026.
- [Screenshots and provenance](docs/screenshots/README.md).*
- ## From “what just happened?” to something you can inspect
- - **Follow the agent behind the activity.** Explore detected processes, parent/child trees, file observations and TCP endpoints in one place.
- - **Investigate sensitive activity.** Open an alert, examine its evidence and attribution, then keep a report for later review.
- - **Check a project before the next run.** Review a selected project, skill or agent profile for static findings, configuration evidence and changes since a saved snapshot.
- - **Understand Claude Code usage.** Inspect measured token subtotals from matching local sessions and estimated costs alongside resource charts.
- - **Add controls where you choose.** Configure explicit action policies and validating MCP gateways for selected routes. [Scope and setup](#selected-action-controls).
- Monitoring runs without an agent plugin. It observes and records; automatic
- blocking requires a separately configured supported route. Recognition and
- attribution can be incomplete. Windows is the primary platform; macOS and Linux
- remain experimental. [Capabilities and limits](docs/CAPABILITIES.md).
- [Install](#install) · [Use AEGIS](#use-aegis) · [FAQ](docs/FAQ.md) · [Roadmap](#current-development) · [Documentation](docs/README.md) · [Contribute](#contribute)
- > **Current source version**: 0.25.0-beta <!-- x-release-please-version -->
- > Each installer contains the features at its release tag.
- > [Verify your download](docs/RELEASE-VERIFICATION.md) · [Beta readiness](docs/BETA-READINESS.md)
- ## 0.25.0-beta highlights
- The [Windows beta published on 12 November 2026 (UTC)](https://github.com/antropos17/Aegis/releases/tag/aegis-v0.25.0-beta)
- includes a signed installer manifest and these improvements:
- - **Choose from eight interface languages.** English, Brazilian Portuguese, Spanish, French, German, Japanese, Simplified Chinese and Russian are available in Settings. The six new catalogs use machine translations with targeted corrections; missing messages fall back to English. [Language details](docs/FAQ.md#which-interface-languages-are-available).
- - **Read translated anomaly notifications.** Anomaly toasts now use the interface translation catalog, with English fallback for missing messages.
- - **Read captured results in your selected locale.** Local security, imported-result and container-capture timestamps follow the selected interface language.
- The release retains safer JSON/CSV export replacement, explicit pricing uncertainty,
- Local security result comparison and the explicitly configured Windows owner-console
- route. [Release notes](https://github.com/antropos17/Aegis/releases/tag/aegis-v0.25.0-beta)
- also list the runtime and development dependency updates.
- The [readiness record](docs/BETA-READINESS.md) distinguishes release integrity,
- source checks and the runtime scenarios tested on specific earlier packages.
- ## Start here
- For a first desktop session: install a verified beta, open **Monitoring** while
- your coding agent runs, then select an agent to inspect its observed activity.
- For a review before running unfamiliar code, open **Local security** and select
- the project folder. No provider API key is needed for these workflows.
- | Your next step | Where to go |
- | --- | --- |
- | Explore the interface | [View screenshots](docs/screenshots/README.md) without installing, or [run the browser preview](#browser-preview) from source. Both use simulated data. |
- | Use the Windows desktop app | [Choose a published release](https://github.com/antropos17/Aegis/releases) and [verify its installer](docs/RELEASE-VERIFICATION.md), then follow [the desktop workflows](#use-aegis). |
- | Help improve AEGIS | [Find a contribution by skill](CONTRIBUTING.md#choose-a-task), check [active PRs](https://github.com/antropos17/Aegis/pulls) for overlap, or [ask a question](https://github.com/antropos17/Aegis/discussions). |
- <a id="capabilities-at-a-glance"></a>
- <a id="what-aegis-observes"></a>
- <a id="monitor-first"></a>
- <a id="observe-your-agents"></a>
- <a id="review-before-use"></a>
- <a id="control-selected-actions"></a>
- ## What AEGIS does
- | Area | Coverage |
- | --- | --- |
- | Processes | 112 agents (265 process-name signatures), process trees, IDE-host and extension detection; Windows primary, macOS/Linux experimental |
- | Files | Changes in configured sensitive directories and agent config paths; Windows open-handle and Restart Manager observations |
- | Network | TCP endpoints for detected agent PIDs, endpoint names and `allowlisted` / `unknown` / `flagged` verdicts |
- | Behavior | 73 sensitive-path detection rules across 8 categories, rolling 10-session baselines, anomaly scoring and sequence evidence |
- | Resources and usage | CPU/RAM charts, local LLM probes, Claude Code token subtotals and estimated costs |
- | Runtime inventory | Running WSL distribution names and local Docker/Linux-native Podman candidates; explicit freshness, no guest process inspection |
- | Local review | Project, skill and profile inventory; static findings, saved snapshot comparison and offline report import |
- | Investigation | Agent instances, file/network views, alerts, audit history, statistics and JSON/CSV/HTML/ZIP exports |
- Monitoring requires no agent plugin or subscription. Default monitoring observes
- and logs; it does not automatically block commands or contain agents. Kill,
- suspend and resume are manual actions. See the [full capability guide](docs/CAPABILITIES.md).
- <a id="download"></a>
- ## Install
- <a id="windows-installer"></a>
- ### Windows
- Download the installer, `manifest.json` and `manifest.json.sig` from the same
- [GitHub release](https://github.com/antropos17/Aegis/releases), then follow
- [offline verification](docs/RELEASE-VERIFICATION.md). Release notes describe that
- installer's features.
- Install the first beta manually after verification: the published 0.17.0-alpha
- updater cannot discover beta releases. When upgrading from 0.14.1-alpha or older,
- install 0.15.0-alpha or newer manually before using signed in-app updates.
- ### From source
- Requires Node.js 24.16.0 or later in the 24.x line. Source uses Electron 45 and
- 64-bit platforms; macOS requires version 13 or later. Windows 10/11 is primary;
- macOS/Linux remain experimental.
- ```bash
- git clone https://github.com/antropos17/Aegis.git
- cd Aegis
- npm ci
- npm start
- ```
- On Windows, run `npm run build:sidecar` after `npm ci` for native Job/MCP routes
- and native-helper tests. Linux process actions additionally require Python 3.9+
- at /usr/bin/python3 and kernel pidfd support; see [Linux process control](docs/development/linux-process-control.md).
- The [development workflow](docs/development/workflow.md) covers helper builds and checks.
- <a id="try-without-ai-agents"></a>
- ### Browser preview
- From a source checkout after `npm ci`, run:
- ```bash
- npm run dev
- # open http://127.0.0.1:8770
- ```
- The preview shares desktop components and uses simulated data with an isolated
- host. It disables native exports and provider requests. `npm run frontend:build:preview`
- builds that preview; `npm run build:renderer` builds the desktop renderer.
- <a id="start-with-a-task"></a>
- ## Use AEGIS
- Source opens **Advanced** navigation by default. **Simple interface** in Settings
- offers investigation, file checks and settings; a saved Simple choice is retained.
- | Task | Start here |
- | --- | --- |
- | See agents and their activity | **Monitoring**, then an agent's processes, files or connections |
- | Review a sensitive-file alert | **Alerts**, then [inspect the captured evidence](docs/ALERTS-AND-PERMISSIONS.md) |
- | Inspect a Windows file read | **Events** → **Folder to monitor** → **Start file read monitoring**; [setup and limits](docs/development/ordinary-file-reads.md) |
- | Review a project, skill or profile | [**Local security**](docs/LOCAL-SECURITY-UI.md) |
- | Check a selected action against a policy | [**Action control**](docs/ACTION-COVERAGE-UI.md) |
- | Review a separate Claude Code hook profile | **Action control → Separate Claude hook profile**; [setup, supported launch and removal](docs/CLAUDE-HOOK-APP.md) |
- | Configure selected-action MCP tools | [Terminal setup](docs/ACTION-MCP-CONFIG.md) |
- | Find other tools | **Start here** or **Commands** (Ctrl K); [workspace guide](docs/OBSERVATORY-GUIDED-WORKFLOWS.md) |
- | Monitoring | Local security |
- | --- | --- |
- | [](docs/screenshots/01-monitoring.png) | [](docs/screenshots/04-local-security.png) |
- These are simulated source UI captures from 11 October 2026.
- [All screenshots and provenance](docs/screenshots/README.md).
- Windows x64 packages include experimental Kernel-File reads. In **Events**, choose
- a local folder and explicitly start capture with manual UAC consent. It stays off
- on startup and stops on Stop, monitoring pause, suspend or quit. Events can display
- a read, its verified issuing agent and
- **Potential sensitive read** from path rules. Paths remain candidates, and the
- label does not establish malicious intent or change risk scores.
- Monitoring data stays local, with no usage telemetry or cloud sync. Endpoint
- naming uses DNS; update checks contact GitHub. Optional Anthropic analysis sends
- activity metadata on request using your own API key and incurs that service's
- charges. See [privacy and key handling](SECURITY.md#privacy-architecture).
- <a id="opt-in-action-control"></a>
- <a id="selected-action-controls"></a>
- ## Selected action controls
- These routes require explicit setup and cover selected actions only. They do not
- intercept other agent tools, arbitrary MCP traffic or activity outside the route.
- | Route | Scope |
- | --- | --- |
- | [Exact execution policy and terminal review](docs/ACTION-EXECUTION.md) | One selected CLI action; policy deny and failed preparation do not launch. [Fresh confirmation](docs/ACTION-CONFIRMATION.md) can be required per attempt |
- | [Windows Job lifetime](docs/ACTION-EXECUTION.md#opt-in-windows-job-lifetime-route) | Confirmed cleanup of ordinary Job-member descendants; caller file/network privileges are retained |
- | [Windows AppContainer](docs/ACTION-APPCONTAINER.md) | One reviewed offline action in a new retained workspace, zero capabilities, private-file access checks and Job cleanup |
- | [Selected-action MCP tools](docs/ACTION-MCP-CATALOG.md) | Up to eight fixed argument-free tools, optional [review broker](docs/ACTION-MCP-REVIEW.md), [configuration checks](docs/ACTION-ROUTE-CHECK.md) and [connection status](docs/ACTION-MCP-STATUS.md) |
- | [Validating MCP gateways](docs/MCP-STDIO-GATEWAY.md) | Explicit stdio, [loopback HTTP](docs/MCP-HTTP-GATEWAY.md) and [pinned HTTPS](docs/MCP-HTTPS-GATEWAY.md); accepted catalogs and bounded grants |
- | [Reviewed file deletion](docs/ACTION-DELETE-FILE.md) | One exact AEGIS-owned file operation with mandatory terminal confirmation |
- Optional [persistent grants](docs/MCP-DURABLE-GRANTS.md),
- [known-secret checks](docs/MCP-KNOWN-SECRETS.md) and
- [live route observation](docs/ACTION-LIVE-OBSERVATION.md) have separate contracts.
- Configuration checks grant no execution permission; observation does not verify
- provider identity or blocking. Client settings stay under operator control.
- Terminal previews and generated configurations can contain private paths or
- secrets, so keep them private. [Setup guides](docs/README.md#configure-selected-actions).
- ## Experimental protected Windows guests
- Administrative repository APIs can create a private Hyper-V VM, hydrate its
- original disk from a trusted Windows VHDX, and run one bounded executable as a
- fresh standard guest user. Actual boot, quota refusal, descendant cleanup and VM
- removal passed on the [recorded source](docs/roadmap/enforcement-e2/protected-guest-hosted-evidence-20261016.md).
- The guest has no NICs, a private desktop, a 512 MiB allocation quota, a
- 32-process/512 MiB Job and a 120-second run limit. These require an administrative
- disposable lab. Ordinary installed-caller launch stays disabled
- (`launchAllowed=false`); complete session integration and independent A1–A4
- acceptance remain open. [Administrative contract](docs/roadmap/enforcement-e2/protected-guest-start.md)
- · [Protection roadmap](docs/roadmap/ai-agent-protection.md).
- <a id="latest-source"></a>
- ## Current development
- The immediate target is a usable Windows Claude Guard: install, launch the
- supported route, allow ordinary work, block declared dangerous attempts and
- inspect the result. This is an active delivery target; full installed-route
- acceptance remains open. See [the execution order](docs/roadmap/completion-plan.md#7-execution-order)
- and [current evidence and gaps](docs/roadmap/guard-delivery-review-20261027.md).
- The [completion plan](docs/roadmap/completion-plan.md) tracks the current work and
- its acceptance boundaries. [Source changes](CHANGELOG.md) and the
- [development context](docs/development/current-context.md) identify the active
- checkout and evidence. Each qualification record names its own tested revision
- and package; a source change does not update a published installer.
- The published [0.25.0-beta release](https://github.com/antropos17/Aegis/releases/tag/aegis-v0.25.0-beta)
- includes six additional language catalogs, translated anomaly notifications and
- dependency updates. Its interface offers eight languages; the six new catalogs
- still benefit from native-language review. See the [language FAQ](docs/FAQ.md#which-interface-languages-are-available)
- for English fallback behavior. Follow [active PRs](https://github.com/antropos17/Aegis/pulls)
- for proposed work; installed features remain tied to the release tag.
- The following is a historical snapshot from 18 October 2026, through
- [#975](https://github.com/antropos17/Aegis/pull/975) at
- [`96b63b16`](https://github.com/antropos17/Aegis/tree/96b63b165ac932e9615815f93c587765e6bde1b8).
- <details>
- <summary>Read the engineering milestones and dated native qualification evidence</summary>
- | Area | Merged work |
- | --- | --- |
- | Development context | [#960](https://github.com/antropos17/Aegis/pull/960) aligns context and skills; [#962](https://github.com/antropos17/Aegis/pull/962) adds Windows fixture CI. [Contributor workflow](docs/development/workflow.md) |
- | Packaged monitoring | [#963](https://github.com/antropos17/Aegis/pull/963) checks the built window and Start/Query/Stop; [#964](https://github.com/antropos17/Aegis/pull/964) fixes startup freshness. [Evidence boundary](docs/development/packaged-desktop-milestone.md) |
- | File and TCP sensors | [#965](https://github.com/antropos17/Aegis/pull/965) refreshes TCP observations after settings changes and checks owned file/socket activity. [Packaged journey](docs/development/packaged-desktop-milestone.md#packaged-file-and-network-sensors) |
- | Diagnostic ETW | [#966](https://github.com/antropos17/Aegis/pull/966)–[#973](https://github.com/antropos17/Aegis/pull/973) test scoped reads and recovery after helper, main and custodian loss. [Recovery limits](docs/roadmap/etw-durable-recovery.md) |
- | Genuine agent read | [#975](https://github.com/antropos17/Aegis/pull/975) verifies installed Claude Code reading a synthetic .env after a preparatory README read, with a visible read, issuer and potential risk. [Live record](docs/recon/evidence/agent-sensitive-read.json) |
- [#974](https://github.com/antropos17/Aegis/pull/974) was closed without merging.
- Its preflight did not establish an actual Windows restart and post-restart capture;
- see the current [Windows recovery admission](docs/roadmap/windows-recovery-admission.md).
- [#976](https://github.com/antropos17/Aegis/pull/976) prepares detected threads
- before the first scoped read. Its [live qualification](docs/recon/evidence/agent-first-read-preparation.json)
- passed the genuine first-read journey without README warmup and the persistent-read
- regression on one diagnostic Windows package. Both showed a verified agent read
- and potential risk; additional observations can remain unverified. This candidate
- was tested with the diagnostic profile. Ordinary packaged controls have their own
- [setup and qualification](docs/development/ordinary-file-reads.md).
- The [idle first-read qualification](docs/recon/evidence/agent-idle-first-read.json)
- passes genuine Claude Code reads after 65, 70 and 85 seconds in an ordinary package,
- with verified issuer, visible action/potential risk and cleanup. It covers renewed
- original handles; the [earlier attribution failure](docs/recon/evidence/ordinary-agent-sensitive-read.json)
- is retained. Collector drops remain separate from kernel loss; these passes do
- not establish complete capture or attribution of every worker.
- </details>
- <a id="the-evidence-graph"></a>
- ## Known limits
- - **Coverage:** Unknown signatures and short-lived processes can be missed. Default monitoring does not parse individual tool calls. Healthy sensors or a valid audit chain do not prove complete capture.
- - **Platform identity:** macOS lacks a process-birth witness. Linux depends on accessible /proc data; fallback lacks a start-time witness. Missing identity limits attribution and process actions.
- - **File evidence:** Candidate paths do not resolve reparse targets. A first read, new worker or unavailable native witness can remain unverified. Diagnostic reads do not enter risk scoring, FileEvents, audit, baselines or sequences.
- - **History and recovery:** UI windows are bounded; Statistics exposes eviction counts and Audit retains persisted records. File-watch retries and ETW recovery have bounded, scenario-specific limits; observation gaps remain explicit.
- - **Token usage:** Claude Code requires a matching process lifetime and readable session data. Bounded reads, oversized records and index capacity can leave subtotals incomplete; other-agent values and dollar amounts can be estimates. [Accounting and soak limits](docs/development/token-accounting-soak.md)
- - **Privacy:** Paths, agent names and endpoints can appear in logs and exports. API-key storage depends on a secure safeStorage backend; [security guidance](SECURITY.md) explains the boundary.
- - **Unmeasured behavior:** No overall detection rate, false-positive rate, startup-time guarantee or whole-app overhead figure has been established. Static findings and scores do not establish safety.
- <a id="development-and-roadmap"></a>
- ## Contribute
- Help complete the Guard journey through setup documentation, reproducible
- failures, meaningful regression cases and decision feedback. The
- [contribution guide](CONTRIBUTING.md#current-product-priority) gives bounded
- starting points and their acceptance conditions.
- Choose an [open good first issue](https://github.com/antropos17/Aegis/issues?q=is%3Aissue%20is%3Aopen%20no%3Aassignee%20label%3A%22good%20first%20issue%22)
- for a small, scoped change, or browse [help wanted](https://github.com/antropos17/Aegis/issues?q=is%3Aissue%20is%3Aopen%20no%3Aassignee%20label%3A%22help%20wanted%22)
- for work that needs more context. The [task picker](CONTRIBUTING.md#choose-a-task)
- connects documentation, Svelte/TypeScript UI and JavaScript/Vitest regression work
- to starting files and focused issue queues. Read the issue's expected result and
- platform needs before starting; small contributions do not require provider
- credentials.
- Start with [CONTRIBUTING.md](CONTRIBUTING.md), the
- [current development context](docs/development/current-context.md) and
- [code conventions](CODING_STANDARDS.md). Main uses CommonJS JavaScript; the Svelte
- renderer and shared types use TypeScript. Exact versions and scripts are in
- [package.json](package.json).
- Use an isolated branch, preserve existing work and describe actual validation in
- your PR. `npm test` runs the configured suite; `npm run counts:check` checks
- maintained declarations. CI also runs builds, lint, type checks, dependency audit
- and derived counts. Mutation campaigns are not part of routine CI. Native and
- provider scenarios need their own matching evidence. [Architecture](ARCHITECTURE.md)
- · [Roadmap](ROADMAP.md).
- <a id="contributors"></a>
- <a id="support-and-license"></a>
- ## Community and license
- **Help more developers see what their agents are doing.** If AEGIS is useful to
- you, star the repository, share a workflow that helped, or tell us what made
- setup difficult in [Discussions](https://github.com/antropos17/Aegis/discussions).
- Reproducible feedback helps shape the next beta.
- For setup questions, start with the [FAQ](docs/FAQ.md) and
- [user guides](docs/README.md), then ask in Discussions. For a reproducible bug,
- use the bug-report form and include your AEGIS version, whether you use an
- installer/source/preview, your OS and the affected workspace. Remove private
- paths, file contents and credentials before sharing logs or screenshots.
- [Bug report](https://github.com/antropos17/Aegis/issues/new?template=01-bug-report.yml)
- · [Feature request](https://github.com/antropos17/Aegis/issues/new?template=02-feature-request.yml)
- · [New agent submission](https://github.com/antropos17/Aegis/issues/new?template=03-new-agent.yml)
- · [Discussions](https://github.com/antropos17/Aegis/discussions)
- · [Private vulnerability report](https://github.com/antropos17/Aegis/security/advisories/new)
- [MIT license](LICENSE) · [Code of conduct](CODE_OF_CONDUCT.md) ·
- [All contributors](https://github.com/antropos17/Aegis/graphs/contributors).
- Navigation icons use a curated Tabler subset with its
- [MIT notice](frontend/observatory/vendor/tabler-icons.LICENSE) and
- [pinned provenance](frontend/observatory/vendor/tabler-icons.provenance.json).
- ## Star history
- [](https://www.star-history.com/?repos=antropos17%2FAegis&type=timeline&legend=top-left)
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.