Context Health Check

    1

    Instant two-line pulse check: context fill level, bloat count, and one recommended action.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Context Health Check

    Example session with this skill installed

    /context-health — quick check before I start this refactor.

    • Read your context and instructions
    • Compiled the context health check

    Crowded session:

    🟠 71% full (~142,000 / 200,000 tokens) — Context is getting crowded — 3 bloat item(s) detected
    💡 Run /find-bloat to identify what to remove.
    

    Fresh session:

    🟢 18% full (~36,000 / 200,000 tokens) — Context is healthy — no bloat detected
    💡 You're good — no action needed.
    

    That's the whole output, by design: two lines, so it doesn't interrupt your flow. For a full breakdown, follow up with Context Window Inspector or Context Bloat Finder.

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

    About this skill

    Returns two lines — a fill-level status with colour-coded emoji and a bloat count, plus one specific recommended action. No tables, no headers, no interruption to your flow.

    LevelEmojiStatus
    0–40%🟢Context is healthy
    41–65%🟡Context is filling up
    66–80%🟠Context is getting crowded
    81–90%🔴Context is nearly full
    91–100%🚨Context is critical

    When to use

    • At the start of any task, as a quick sanity check
    • Periodically during long sessions to catch fill-up early
    • Anytime you want a fast answer without a full report

    How it works

    Invoke it by name in your agent (in Claude Code: /context-health). It reads only what is already in the current session's context — no files are written, no network calls, no external services.

    Part of Claude Token Inspector, an open-source (MIT) suite of 8 context-window skills by the original author. Source: https://github.com/msishariful/claude-token-inspector

    Token counts are estimates (1 token ≈ 4 chars). Default context limit is 200k tokens unless otherwise specified.

    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 any payment.

    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

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 23 days ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed23 days ago

    What's inside

    Frequently Asked Questions

    Context Health provides an instant two-line summary of your current token usage and identifies 'bloat'—unnecessary files or history—that could lead to truncated responses or lost logic during complex tasks.

    It's a standard SKILL.md skill, built and tested on Claude Code. It also works in other SKILL.md-compatible agents (Cursor, Codex CLI, Gemini CLI, etc.). Token counts use a ~4-characters-per-token estimate.

    There's no setting to change. It assumes a 200,000-token window unless there's evidence in the session of a different limit, in which case it uses that.

    The SKILL.md instructions file for the health check. It's free and open source (MIT); the rest of the Claude Token Inspector suite (including Context Bloat Finder) is listed separately and on GitHub.

    No. The skill instructions are short and the output is two lines, so running it adds very little to your context.

    New in LLM Ops