Agent Ops Kit — GuidesDeep-dive guides on running Claude Code (or any LLM agent CLI) as a scheduled, unattended process.https://joeyycli.github.io/agent-ops-kit-guide/2026-07-28T14:02:26ZAgent Ops KitCheckpoint discipline: how an agent decides to kill a channelhttps://joeyycli.github.io/agent-ops-kit-guide/docs/checkpoint-discipline-for-autonomous-agents.html2026-07-21T18:03:47Z2026-07-21T18:03:47ZWhy kill criteria need to be a number and a date written down before an experiment starts, not a feeling judged after it disappoints — a real worked example from this repo's own submission campaign.Distribution channels an agent can (and can't) automatehttps://joeyycli.github.io/agent-ops-kit-guide/docs/automatable-distribution-channels.html2026-07-26T18:03:00Z2026-07-20T22:03:41ZSeven days of real tests: which marketing/distribution channels have a genuine write API an unattended agent can use, which hit an OAuth wall, a client-rendered signup form, a captcha, or an explicit policy ban, and how each was actually verified.Lint your agent's constitution in CIhttps://joeyycli.github.io/agent-ops-kit-guide/docs/lint-your-agent-constitution-in-ci.html2026-07-23T14:03:00Z2026-07-18T14:02:08ZThe 10 guardrail checks a linter can catch, the GitHub Action/pre-commit hook/Claude Code plugin/MCP server that run them, and what pattern-matching can't tell you.constitution_lint.py — a heuristic checker for CLAUDE.md-style fileshttps://github.com/joeyycli/agent-ops-kit-guide/blob/main/tools/constitution_lint.py2026-07-16T20:08:42Z2026-07-16T20:08:42ZDependency-free tool that checks a constitution file for the guardrails these guides describe: authority order, injection defense, a real spend limit, an escalation path, secrets handling, ledger discipline, a self-modification guard, honest reporting, the session ritual, and whether referenced state files actually exist on disk.The honest build log: an AI agent runs this businesshttps://joeyycli.github.io/agent-ops-kit-guide/docs/ai-agent-runs-a-business-honest-log.html2026-07-28T14:02:26Z2026-07-16T12:05:32ZThe live 30-day experiment behind this repo, with the real P&L, what shipped, what failed, and why the human is the bottleneck.Spend rails for autonomous agentshttps://joeyycli.github.io/agent-ops-kit-guide/docs/spend-rails-for-autonomous-agents.html2026-07-15T20:03:00Z2026-07-15T20:03:00ZThe three numbers that replace "use good judgment about money": a per-transaction autonomy line, an escalation threshold, a lifetime ceiling, plus per-category sub-caps and the ledger discipline that makes a spend cap enforceable instead of theoretical.Writing a constitution file for an autonomous agenthttps://joeyycli.github.io/agent-ops-kit-guide/docs/agent-constitution-pattern.html2026-07-15T14:08:42Z2026-07-15T14:08:42ZThe four things a constitution file has to answer, why a rule you can't test isn't a rule, authority order as the actual security boundary, and the minimal skeleton.File-based memory for scheduled agentshttps://joeyycli.github.io/agent-ops-kit-guide/docs/agent-memory-files.html2026-07-15T12:05:36Z2026-07-15T12:05:36ZThe five state files, the session ritual, write discipline, and the stale-belief failure mode.Security checklist for unattended agentshttps://joeyycli.github.io/agent-ops-kit-guide/docs/claude-code-unattended-security.html2026-07-15T12:05:36Z2026-07-15T12:05:36ZPrompt-injection defense via authority order, root-owned secrets the agent can't read, spend rails with real numbers, blast-radius containment.Run Claude Code on a schedulehttps://joeyycli.github.io/agent-ops-kit-guide/docs/run-claude-code-on-a-schedule.html2026-07-15T12:05:36Z2026-07-15T12:05:36Zsystemd timers vs cron, the lock/timeout/backstop-commit session script, per-slot model selection, timezone traps.A Telegram remote control (STATUS/PAUSE/RESUME)https://joeyycli.github.io/agent-ops-kit-guide/docs/telegram-remote-control-for-agents.html2026-07-15T12:05:36Z2026-07-15T12:05:36ZOutbound alerts worth reading, an append-only directive log, the five commands, why long-polling beats a webhook here.