Trying To Reduce Codex Token Burn Without Adding More Noise
A measured Codex token-use note: Token Frugal helped on a small repo, but my rule-heavy setup exposed the cost of adding another instruction layer.
Agents
CodextokensagentsToken Frugal
Chronological archive
Journal
Everything is here, from long experience notes to shorter findings. A chronological archive for experiments, readings, and decisions worth keeping.
All posts
10 published entries.
A Codex plugin for testing dynamic workflows
A bounded Codex Dynamic Workflows lab with a local runtime, MCP server, skill, plugin packaging, real read-only campaigns, and a pause after token-heavy testing.
Agents
CodexMCPClaudeworkflowplugin
Installing Hermes with Codex, almost without touching the keyboard
A local Hermes Agent installation driven by Codex: Open WebUI, OpenAI-compatible API, launchd, streaming errors, and filesystem guardrails.
Agents
CodexHermesOpen WebUImacOSlaunchd
Does Stax really improve an established process?
An SEO lab with four Codex variants to see whether Stax helps an already-framed workflow, and whether a skill makes the result more reliable.
Agents
StaxCodexskillsSEObenchmark
Helping a friend drive Pipedrive with Claude
An MCP lab around Pipedrive: start from a concrete need, refuse the all-powerful connector, then validate real CRM workflows on disposable data.
Agents
MCPPipedriveClaudeCodexCRM
When Codex slows down, look at ~/.codex first
A local investigation into perceived Codex slowness: SQLite logs, WAL files, archived JSONL sessions, and a small macOS maintenance script.
Automation
CodexSQLitemacOSlaunchdmaintenance
AGENTS.md, skills, and harness: where should the rules live?
A practical question about coding agents: which instructions should stay next to the project, and which ones deserve to become reusable?
Agents
CodexskillsAGENTS.mdharnessworkflow
LLM Field Notes: keeping useful words for agents
An annotated reading of LLM Field Notes, useful for separating precise vocabulary from labels that merely sound technical.
Reading
LLMglossaryagentsreading
Exposing a local service without pretending to know
Comparing Tailscale Funnel, Cloudflare Tunnel, and hosting alternatives without turning an intuition into a recommendation.
DevOps
TailscaleCloudflarecontainersdnstunnel
Cursor OpenRouter Proxy
A Go proxy that routed Cursor through OpenRouter, tested DeepSeek, and made Cursor's model payloads inspectable.
AI tooling
CursorOpenRouterproxyagents