Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add KirKruglov/claude-skills-kit --skill context-window-health-checkgit clone --depth 1 https://github.com/KirKruglov/claude-skills-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kirkruglov/claude-skills-kit/context-window-health-check)<a href="https://agentmods.dev/skills/kirkruglov/claude-skills-kit/context-window-health-check"><img src="https://agentmods.dev/badge/skills/kirkruglov/claude-skills-kit/context-window-health-check/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kirkruglov/claude-skills-kit/context-window-health-check"><img src="https://agentmods.dev/badge/skills/kirkruglov/claude-skills-kit/context-window-health-check.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.01301 |
| Opus 5 | $0.00051 | $0.00651 |
| Sonnet 5 | $0.00020 | $0.00260 |
| Haiku 4.5 | $0.00010 | $0.00130 |
Grade A, and why
context-window-health-check scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Window Health Check
This skill assesses the health of the current Claude session for non-technical users and delivers a plain-language status with a single, concrete recommendation: keep going, start a handoff, or open a new session. It works entirely in-chat — no token counts, no code, no external tools.
Input: The user's request (any phrasing); the current session conversation (length, topic diversity, symptom descriptions)
Output: A structured in-chat response — status indicator (🟢 / 🟡 / 🔴), plain-language explanation, one concrete recommendation, and an optional next-step suggestion.
Language Detection
Detect the user's language from their message:
- If Russian (or contains Cyrillic): respond in Russian
- If English (or other Latin-script language): respond in English
- If ambiguous: respond in the language of the trigger phrase used
Instructions
Step 1: Detect Trigger Intent
- Identify whether the user is explicitly requesting a context check or implicitly signalling a problem:
- Explicit: "check my context", "context health check", "is my context running out", etc.
- Implicit: "Claude, you seem to be forgetting…", "you said something different earlier", "this session is getting long"
- In both cases proceed to Step 2.
- If the message is unrelated to session state (e.g., "write me an email") — do not activate this skill; hand off to standard Claude behaviour.
Step 2: Read Session Signals
Analyse the visible conversation to collect the following signals:
-
Message count — count visible user + assistant turns:
- < 10 turns → low pressure
- 10–30 turns → moderate pressure
-
30 turns → high pressure
-
Topic diversity — count distinct topic areas in the conversation:
- 1 topic → cohesive, lower risk
- 2–3 topics → moderate branching
- 4+ topics → high branching, elevated risk
-
Repetition or contradiction signals — check whether the user has mentioned:
- Claude forgetting earlier instructions
- Contradictory answers across the session
- Having to repeat context already provided
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 126 lines · 102 tokens per session scan A 8db0d76680f9
context-window-health-check is a skill published in the GitHub repository KirKruglov/claude-skills-kit (18 stars, last pushed 1mo ago), licensed MIT. It adds 102 tokens to every session and 1,301 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
memory-management
Two-tier memory system that makes Claude a true workplace collaborator. Decodes shorthand, acronyms, nicknames, and internal language so Claude understands requests like a colleague would. CLAUDE.md for working memory, memory/ directory for the full knowledge base.
ai-agents-architect
Expert in designing and building autonomous AI agents. Masters tool use, memory systems, planning strategies, and multi-agent orchestration.
context-engineering
A project documentation framework that records a project's goals, requirements, roadmap, and current state in separate files. It also keeps those documents within set size limits for an AI coding agent.
ask
Adaptive vault reader skill. Receives a natural language question, searches the vault first (Glob/Grep, entity reads, wikilink traversal), then self-assesses whether more context is needed. Escalates to /graphify for graph-level understanding or to /bedrock:teach for remote content ingestion only when the vault alone…
compress
Vault alignment engine. Detects and fixes 5 types of structural misalignments: broken backlinks, concept fragmentation, entity miscategorization, duplicated entities, and misnamed entities. Delegates all writes to /bedrock:preserve. Supports interactive mode (user confirmation) and cron mode (autonomous mechanical…
waggle-recall
Recall project-scoped Waggle history about a topic, component, decision, bug, experiment, or implementation attempt.