generate-week-x
Skill0 starsChanged 2 months agoMITView on GitHub
Use when the user asks to plan and generate this week's X (Twitter) posts end to end, wants the full weekly X pipeline run in one go, or explicitly invokes /generate-week-x. Plans the week's X angles from the shared idea pool, then generates each post/thread through its own isolated subagent, checks each against the week's X history, then runs one combined review + scheduling pass. Never pads the count to hit a quota.
In this repository38 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- audit-draftSkill
- check-plagiarismSkill
- critique-draftSkill
- critique-draft-substack-articleSkill
- critique-draft-substack-noteSkill
- critique-draft-xSkill
- draft-commentSkill
- draft-replySkill
- extract-hookSkill
- generate-ideasSkill
- generate-visualSkill
- generate-visual-substack-articleSkill
- generate-weekSkill
- generate-week-substackSkill
- humanize-draftSkill
- interviewerSkill
- monitor-engagementSkill
- optimize-profileSkill
- plan-advocacySkill
- plan-weekSkill
- plan-week-substack-articleSkill
- plan-week-substack-noteSkill
- plan-week-xSkill
- publish-substackSkill
- pull-analyticsSkill
- pull-analytics-xSkill
- repurpose-postSkill
- research-topicSkill
- review-draftsSkill
- schedule-approvedSkill
- schedule-approved-xSkill
- update-playbookSkill
- write-draftSkill
- write-draft-substack-articleSkill
- write-draft-substack-noteSkill
- write-draft-xSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
Run, in order, via the Skill tool:
- Very specific: line 34
If any slot is unfilled, run one targeted /research-topic <pillar> 2 → /generate-ideas → /plan-week-x retry, capped at 2 extra rounds.
- Very specific: line 72
After both, update content-index.md's status column for each X row.
- Very specific: line 82
Never skip /review-drafts before /schedule-approved-x.
- Distinctive: line 83
Never exceed 2 retries per slot.
- Distinctive: line 85
Never pad the week's count.
- Very specific: line 86
If /schedule-approved-x's required env vars are missing, stop at that step and report exactly what's missing.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Generate Week X (Weekly Orchestrator — X pipeline)
/generate-week's sibling for X, same per-post-subagent isolation pattern,
default target 5 (per REQUIREMENTS.md's X cadence). Reads from the same
shared Post-Ideas/ pool /generate-week also draws from — an idea
tagged for both x and linkedin can be picked up by both orchestrators
independently without conflict, since each platform claims its own
platform_schedule entry rather than overwriting a shared one.
Arguments
- week (optional) — passed through to
/plan-week-x. - target_count (optional) — passed through to
/plan-week-x. Default 5.
Process
1. Plan the week's X angles
Run, in order, via the Skill tool:
/research-topicwith a broad scan sized for X's higher cadence — start at 10 given more slots need to clear dedup/verification than LinkedIn's weekly 8./generate-ideas(default range) — this is the same shared research/generate-weekalso draws from; don't re-research topics/generate-weekalready covered this week if it ran first — checkPost-Ideas/for recent unclaimed ideas before assuming a fresh research pass is needed./plan-week-x [week] [target_count].
If any slot is unfilled, run one targeted /research-topic <pillar> 2 →
/generate-ideas → /plan-week-x retry, capped at 2 extra rounds.
Accept the honest gap after that.
2. Generate each post independently, one subagent at a time
For each idea now carrying an x entry in platform_schedule, in day
order, spawn one fresh subagent (Agent tool, subagent_type:
general-purpose, run_in_background: false) with:
- This slot's idea id, topic, pillar, angle, hook, sources, and its
xdate fromplatform_schedule. - The topic + pillar only of every other idea assigned to X this week (and, if known, this week's LinkedIn/Substack picks too — cross-platform awareness prevents the account from saying the identical thing everywhere in the same week with zero differentiation).
- The current
Content-Learnings/content-index.mdcontents. - An instruction to read and follow, directly:
.claude/skills/ write-draft-x/SKILL.md, then.claude/skills/critique-draft-x/SKILL.md, then.claude/skills/generate-visual/SKILL.md(platform: x). Point at the real files, don't paraphrase their logic into the prompt. - The same retry-on-duplicate instruction as
/generate-weekstep 2, scoped to the X pool and capped at 2 retries total for the slot. - An instruction to append/update one row in
content-index.mdfor this post (withplatform: x), same lifecycle as/generate-weekdescribes. - An instruction to report back: final status, viral_score, single/thread and tweet count, visual format, and whether it had to re-angle.
3. Combined review and scheduling
/review-draftswith no draft-id — naturally covers everystatus: in_reviewnote regardless of platform, so if/generate-weekalso ran this session, this becomes one shared review pass rather than two separate ones. Run it once total per session, not once per orchestrator./schedule-approved-xonce, for whatever ended upplatform: x,status: approved.
After both, update content-index.md's status column for each X row.
4. Final report
Same shape as /generate-week's: a date|day|topic|category|viral_score|
format table, a week-shape line, a variety check, and final counts —
scoped to this week's X posts.
Hard rules
- Never skip
/review-draftsbefore/schedule-approved-x. - Never exceed 2 retries per slot; never fall back to a weaker/duplicate idea to fill a day.
- Never pad the week's count.
- If
/schedule-approved-x's required env vars are missing, stop at that step and report exactly what's missing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

