This skill ensures that the agent remains within the standard pricing tier (<128k tokens) for Google Gemini models by actively managing context bloat.
Rules of Operation
Always-On Counter: The agent MUST include the current token usage (e.g., (Current Usage: 45k/128k)) at the end of every message while this skill is installed.
Context Monitoring: The agent must check the current token count (via session_status) before performing token-heavy tasks like image analysis.
The 128k Wall: If the current context is above 110,000 tokens, the agent MUST notify the user before the next turn that they are approaching the "Double Billing" zone.
Hard Reset: If the context exceeds 125,000 tokens, the agent will:
Automatically summarize all key facts/decisions from the current session into MEMORY.md.
Request a session reset (/new) or manually trim the history to bring the count back to near zero.
Output Discipline: All responses must be "Logic-First" and concise. No conversational filler or repeated instructions.
When to Use
Triggered automatically when context > 110k.
Can be invoked manually by the user saying "Activate Financial Shield."
1---2name: financial-shield3description: Financial Shield4---56# Financial Shield78This skill ensures that the agent remains within the standard pricing tier (<128k tokens) for Google Gemini models by actively managing context bloat.910## Rules of Operation11121. **Always-On Counter:** The agent MUST include the current token usage (e.g., `(Current Usage: 45k/128k)`) at the end of every message while this skill is installed.132. **Context Monitoring:** The agent must check the current token count (via `session_status`) before performing token-heavy tasks like image analysis.143. **The 128k Wall:** If the current context is above 110,000 tokens, the agent MUST notify the user before the next turn that they are approaching the "Double Billing" zone.154. **Hard Reset:** If the context exceeds 125,000 tokens, the agent will:16 - Automatically summarize all key facts/decisions from the current session into `MEMORY.md`.17 - Request a session reset (`/new`) or manually trim the history to bring the count back to near zero.185. **Output Discipline:** All responses must be "Logic-First" and concise. No conversational filler or repeated instructions.1920## When to Use21- Triggered automatically when context > 110k.22- Can be invoked manually by the user saying "Activate Financial Shield."
Run npx skillmds@latest add kbarbel640-del/financial-shield in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Financial Shield It is listed under Coding & Dev Tools on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
kbarbel640-del (@kbarbel640-del) published this skill. Their other Agent Skills are listed on their SkillMD profile.
kbarbel640-del