More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    ENHE Project Sentinel

    2

    Local-first, read-only tracking for multiple projects with bilingual progress, risk, ETA, and Codex usage reports.

    Free

    2 usesSecurity scanned
    ENHE Project Sentinel

    ENHE Project Sentinel

    Example session with this skill installed

    Install ENHE Project Sentinel under %LOCALAPPDATA%\ENHE\Project-Sentinel, register my three local repositories, and generate today's Chinese and English Markdown and HTML project portfolio reports in read-only mode. Include progress, ETA confidence, risks, quality signals, Codex token usage, and next actions without running project commands.

    • Read your context and instructions
    • Compiled the enhe project sentinel

    Sentinel installed under %LOCALAPPDATA%\ENHE\Project-Sentinel. Three registered repositories were scanned in read-only mode. Chinese and English Markdown and HTML reports were generated.

    Highest risk: Project A has low test coverage.
    ETA: insufficient evidence for a precise estimate.
    Next action: review the authentication module and add unit tests after user approval.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    ENHE Project Sentinel is a local-first, read-only Skill for tracking multiple software and AI projects. It reads explicitly registered project directories, Git history, project documents, development reports, quality signals, and available Codex usage data, then produces bilingual Chinese and English project portfolio reports.

    Reports can include project counts, progress estimates, confidence levels, elapsed development time, ETA ranges, recent activity, branch and workspace changes, test, lint, typecheck, and build signals, risks, blockers, Codex context counts, token usage, next actions, and agent-ready prompts.

    It is designed for indie developers, AI product builders, product owners, and small R&D teams. It runs on Windows 10/11 and does not upload source code or modify tracked project code by default. It also does not run project commands by default. Progress, health, and ETA are evidence-based management indicators, not delivery guarantees. When evidence is insufficient, the report clearly says so instead of inventing a precise ETA.

    Optional features include HTML and JSON reports, SMTP email delivery, and a Windows 08:00 scheduled task. These features require explicit user configuration. The package includes the SKILL.md, installation and runtime scripts, configuration templates, report formats, and a ZIP release package. The Skill and included source resources use the Apache License 2.0.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Business & Operations.

    Windows-first: requires Windows 10/11, PowerShell 5.1+, Node.js 20.19+, npm, and Git.

    Project directories must be explicitly registered. Symbolic links and Junctions are not followed outside registered project roots.

    Default mode is read-only for tracked projects: safeScanOnly=true and allowRunCommands=false.

    Progress, health, confidence, and ETA are evidence-based management indicators, not delivery promises.

    This product is not a cloud SaaS and does not automatically modify code, deploy applications, run migrations, or operate production systems.

    SMTP email and the daily 08:00 scheduled task require explicit configuration and authorization.

    The current release should not be advertised as having verified one-click macOS/Linux support.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    2 uses

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    • Free with an Agensi account

    Accounts
    Gmail
    Software
    Node.js

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Environment Variables

    Allowed Hosts

    smtp.gmail.com

    This Skill reads explicitly registered project directories and available Codex state to generate reports. It writes Sentinel-owned reports, snapshots, knowledge-base files, configuration, and rebuildable runtime files. It does not run tracked-project commands by default. Browser and network access are not required for local reporting. Optional SMTP email requires explicit user configuration and authorization. Project secrets, cookies, tokens, passwords, and credentials must not be included in reports.

    Listed3 months ago

    What's inside

    Frequently Asked Questions

    The Skill provides an evidence-based, bilingual (Chinese/English) overview of your entire project portfolio, identifying risks and ETA ranges based on actual Git history and quality signals rather than manual status updates.

    It is built for Windows 10/11 environments and is designed to integrate with AI agents that can parse local file structures, providing them with 'agent-ready' prompts for further task planning.

    No, the Skill is local-first and read-only; it only accesses directories you explicitly register and does not upload your source code or modify your files by default.

    The package includes the SKILL.md, installation and runtime scripts, configuration templates for SMTP and scheduled tasks, and the full ZIP release package under the Apache License 2.0.

    Unlike tools that guess a date, Sentinel analyzes development velocity and quality signals; if there isn't enough data to make a reliable prediction, it will explicitly state the evidence is insufficient rather than inventing a deadline.

    New in Business & Operations

    Free