lennoxsaint/Threadify-workflows / plugins/claude/skills/threadify-run-my-threads
threadify-run-my-threads
VerifiedSet up one daily schedule so your agent prepares five Threadify-written Threads posts every morning from a weekly plan (one long-form thread on six days, one greatest-hit repost a day, hook-first structures and a chosen CTA level), shows them in this thread, and schedules them only after you reply "yes". Use when someone asks Threadify to run their Threads, prepare posts every morning for approval, or says "I want to approve posts, not write them".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 21.0 KB |
| references/daily-schedule.md | markdown | 6.0 KB |
| references/generation-models.json | config | 932 B |
| references/hook-bank.md | markdown | 976 B |
| references/state-and-recovery.md | markdown | 3.5 KB |
| references/threadify-001.md | markdown | 3.7 KB |
| references/threads-playbook.md | markdown | 2.7 KB |
| references/workflow-manifest.json | config | 4.4 KB |
| references/workflow-readme.md | markdown | 1.0 KB |
| scripts/casing-guard.mjs | script | 6.6 KB |
| scripts/hook-check.mjs | script | 7.2 KB |
| scripts/run-state.mjs | script | 9.5 KB |
| scripts/week-plan.mjs | script | 14.2 KB |
references/daily-schedule.md
View on GitHub# Run My Threads daily schedule
Run My Threads uses exactly one host schedule that fires once a day. The default time is 05:30 in the owner's timezone; the owner may choose another time. The run prepares the approval card and stops. Nothing is scheduled on Threads until the owner replies.
## Find before you create
Before creating anything, list the host's existing schedules and look for one named "Run My Threads" or whose prompt runs `threadify-run-my-threads`.
- One found: reuse it. Show its name, time, timezone, status and next run. Change its time or settings in place only when the owner asks.
- More than one found: show them all and ask which one to keep. Pause or delete the others only with the owner's approval.
- None found: create one, then read it back.
Never create a second schedule to change a time or a setting. Never add an hourly or repeating check.
## The daily prompt
Save the owner's choices in the schedule's prompt so each run is self-contained:
> Run the installed threadify-run-my-threads skill for today. Account: @HANDLE. Timezone: AREA/CITY. Week plan: start_date YYYY-MM-DD, mode growth | balanced | conversion, short_day WEEKDAY, seed N, auto_plug true | false. Offer: OFFER_ID or none. Casing: as Threadify writes it | lowercase except proper nouns, keep: NAMES. Topic lanes: LANES or let Threadify choose from my Brain. Approved destination: DESTINATION. Sources: optional connected Viral Vault, My Vault and own greatest hits. Policy: read the installed SKILL.md, threads-playbook.md and state-and-recovery.md on each run; these own generation, revision and recovery behavior. State folder: ~/.threadify-workflows/state/run-my-threads/. Prepare today's five posts and the approval card in this conversation and wait for my reply. Never schedule without my "yes" here. If today's card or receipt already exists, do not generate again.
The week plan values let any run rebuild the same plan with `scripts/week-plan.mjs`, even a cloud run without the local state folder. When the seven days end, the run moves `start_date` on seven days and adds one to `seed`; update the schedule prompt in place (never a second schedule). The state folder keeps the setup choices, the current week plan and the greatest-hit repost log; without it, the 60-day repost check relies on Threadify's `search_posts` and `list_scheduled_posts`.
A schedule's own prompt is never approval to schedule posts. Only the owner's reply to the card is.
## Codex
Create the schedule with Codex's own automation controls: ask Codex in this thread to add a daily automation, or use the Automations view. Do not hand-write Codex state files.
Prefer a thread automation attached to this thread, so every morning's card arrives in the conversation the owner opens from the Codex app and the "yes" lands in the same place. If thread automations are unavailable, use a project automation with the Threadify connection available; each run then opens a new thread. Codex saves an automation as `automation.toml` in its own automations folder; a thread automation reads back roughly as:
```toml
version = 1
id = "run-my-threads"
kind = "heartbeat"
name = "Run My Threads"
prompt = "Run the installed threadify-run-my-threads skill for today. ..."
status = "ACTIVE"
rrule = "FREQ=DAILY;BYHOUR=5;BYMINUTE=30;BYSECOND=0"
target_thread_id = "THIS_THREAD"
```
A project automation uses `kind = "cron"` with a project target and working folder instead of `target_thread_id`. Read back the name, `ACTIVE` status, daily rule and the next run Codex shows, and confirm that next run is 05:30 (or the chosen time) in the owner's timezone. Pause by setting the automation to paused in Codex.
## Claude Code
Preferred: a cloud routine. Create it with `/schedule` in the Claude Code CLI, at claude.ai/code/routines, or from the Desktop app's Routines page (New routine, Cloud). Choose a daily schedule at 05:30 local; times are entered in local time and converted for you. Read back the next run time.
- Connectors: include the Threadify connector from the owner's claude.ai connectors. A server added only on this computer with `claude mcp add` is not available to a cloud routine. Remove connectors the routine does not need.
- Skill: a routine runs in a fresh cloud copy of the repository you select, not on this computer. Select a repository that contains the installed skill, or the public `lennoxsaint/Threadify-workflows` repository with the prompt telling Claude to read `skills/threadify-run-my-threads/SKILL.md` at the installed release tag.
- Approval from the phone: each run creates a session. Open it in the Claude app and reply there. Routines run without permission prompts, so the skill's own stop-before-scheduling rule is what protects the owner.
- Manage: `/schedule list`, `/schedule update`, the on/off switch to pause.
Fallback when Threadify is configured only on this computer: a Desktop local scheduled task (Routines, New routine, Local, Daily 05:30). It runs only while the Desktop app is open and the computer is awake. A missed run catches up once on wake, so the run must plan slots from the current time, not from 05:30.
Do not use `/loop` or in-session scheduled tasks for this workflow. They need an open session and recurring ones expire after seven days.
## Other hosts
Use the host's one native daily scheduler (Cursor Automations, Gemini CLI, OpenClaw or Hermes cron) with the same prompt, or an owner-approved OS schedule that runs the agent's non-interactive command. Show the exact command, time and timezone and get consent before registering it. The same one-schedule rule applies.
## Usage readout
Each run reports what it used only when the host exposes it (for example a token or cost figure in the run details). Threadify's remaining generation quota comes from `get_connection_defaults` before and after the run. When the host shows no usage figure, say "usage not exposed by this host". Never estimate.
Notify only when a card is ready or an actionable blocker changes. Stay quiet for unchanged state. Expire unapproved prior cards only when preparing a new daily card; never migrate historical card text during a policy upgrade.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 470k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |