# TokenMaxer

TokenMaxer publishes Codex-readable behavior patches, workflow notes, and public AI operating loops.

## Posts

- AI Agent Control Gates: Stop Bad Agents Before They Act: https://tokenmaxer.ai/blog/agent-control-gates.md
- Agent Observability: Trace What Agents Decide and Do: https://tokenmaxer.ai/blog/agent-observability.md
- Agent Testing: Test Plans for Tool-Using AI Systems: https://tokenmaxer.ai/blog/agent-testing.md
- Agent Tracing: A Practical Schema for Tool-Using AI: https://tokenmaxer.ai/blog/agent-tracing.md
- AI Agent Evaluation: Gates That Catch Bad Behavior: https://tokenmaxer.ai/blog/ai-agent-evaluation.md
- AI Agent Governance: Approval, Audit, and Eval Gates: https://tokenmaxer.ai/blog/ai-agent-governance.md
- AI Agent Monitoring: Metrics, Logs, and Stop Conditions: https://tokenmaxer.ai/blog/ai-agent-monitoring.md
- AI Agent Security: Threat Models for Tool-Using Agents: https://tokenmaxer.ai/blog/ai-agent-security.md
- Bad CPA Is Not a Diagnosis: A Constraint Map for Meta Ads: https://tokenmaxer.ai/blog/bad-cpa-is-not-a-diagnosis.md
- Human Approval for AI Agents: When Agents Should Stop: https://tokenmaxer.ai/blog/human-approval-ai-agent.md
- LLM Observability: When Basic Telemetry Stops Working: https://tokenmaxer.ai/blog/llm-observability.md
- Better Embeddings for Vector Search Start Before the Database: https://tokenmaxer.ai/blog/llms-make-better-embeddings.md
- MCP Authentication: Gate Agent Access to Tools Safely: https://tokenmaxer.ai/blog/mcp-authentication.md
- MCP Security: Review Checklist for Agent Tool Servers: https://tokenmaxer.ai/blog/mcp-security.md
- Prompt Caching: Cut Agent Cost Without Breaking Quality: https://tokenmaxer.ai/blog/prompt-caching.md
- Agent Harness Architecture: 15 Jobs Around the Model Loop: https://tokenmaxer.ai/blog/stop-importing-your-agent-harness-build-it.md

## Interfaces

- Raw Markdown: /blog/:slug.md
- Structured catalog: /codex/index.json