plan-week
Skill0 starsChanged 59 days agoMITView on GitHub
Use when the user asks to plan the week's LinkedIn content, decide what gets posted when, run the Content Strategist/Content Planner, or explicitly invokes /plan-week. Produces a 7-day calendar VIEW (Mon-Sun) but only fills the quality-gated slots that clear the bar (default ~3/week, Tue/Thu/Sat starting heuristic) per REQUIREMENTS.md §5 -- remaining days are shown explicitly empty with a stated reason, never padded. Each filled day gets a topic, format, a hook formula (from Content-Learnings/hook-formulas.md), a posting time (per §11, overridden by playbook.md evidence), and 2-3 comment-target suggestions. Does not draft post text (that's /write-draft).
2 other projects use a version of this.
Adopted by
Most recently taken up by
- ozgurkarahan/ai-agent-memorytheir own version, first seen Jul 24, 2026
- aaronwskinner/claude-skillstheir own version, first seen Jun 1, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- ozgurkarahan/ai-agent-memory★ 8same name
- aaronwskinner/claude-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
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
- generate-week-xSkill
- humanize-draftSkill
- interviewerSkill
- monitor-engagementSkill
- optimize-profileSkill
- plan-advocacySkill
- 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
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 4 are distinctive, and 9 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 29
Default 3. This is a ceiling to aim toward, not a quota to force — see the hard rules below.
- Very specific: line 35
Read all Idea Notes in Post-Ideas/ with status: candidate (never placeholder).
- Very specific: line 49
Days: read Content-Learnings/playbook.md first.
- Very specific: line 85
Fit check: prefer a canonical formula whose engagement_goals value matches the idea's likely engagement goal — the same precedence …
- Very specific: line 95
This week's own picks: never assign the same formula_id to two different filled slots within the same plan produced by this run.
- Very specific: line 102
If content-index.md doesn't exist yet or is too thin to judge from, this check is a no-op — don't block on it, same fallback posture as …
- Distinctive: line 105
No genuine fit: if nothing in the canonical set fits this idea's category/content_type (a goal mismatch alone isn't disqualifying if the …
- Very specific: line 124
Override: if Content-Learnings/playbook.md has a Best-Performing Patterns rule with real evidence (≥3 published posts, post ids attached) …
- Distinctive: line 146
Neither source yields anything: leave the field explicitly empty in the report with a one-line reason (e.g.
- Very specific: line 166
If fewer than target_count slots ended up filled because the pool genuinely didn't support more distinct, valuable angles, say that plainly …
- Very specific: line 173
Never assign more than target_count (default 3) posts in one week, and never assign to Sunday unless the user explicitly asks for it this …
- Distinctive: line 179
Never draft post text here — ideas only get a day, not content.
- Very specific: line 180
Never pick a status: placeholder idea.
≈ “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
Plan Week (Content Strategist / Content Planner — Phase 4a, revised Phase 13, reworked Phase 21)
Assigns ranked candidate ideas to specific days under the flexible weekly
cadence (REQUIREMENTS.md §5), then — for each slot that actually gets
filled — attaches a suggested hook formula, a posting time, and 2-3
comment targets (REQUIREMENTS.md §32), and renders the whole thing as a
full Mon-Sun calendar view with unfilled days shown explicitly empty
rather than hidden. Selection and scheduling-day decisions only — does
not write post text (that's /write-draft) and does not touch Buffer or
actual scheduling (that's a later phase). §32 is additive to this file's
original §5 logic: it changes nothing about which ideas get picked or how
many slots get filled, only what gets attached to a slot once it's filled
and how the result is displayed.
Arguments
- week (optional) — an ISO week (e.g.
2026-W38) or a Monday date to plan for. If omitted, default to the next full week from today's actual date (if today is on or before Saturday of a week that hasn't started posting yet, that week; otherwise the following week). Never plan into a week that already hastarget_countslots filled withstatus: selectedor later. - target_count (optional) — how many posts to aim for this week. Default 3. This is a ceiling to aim toward, not a quota to force — see the hard rules below.
Process
1. Load the candidate pool
Read all Idea Notes in Post-Ideas/ with status: candidate (never
placeholder). Note each one's category, content_type, format,
rank_score.
2. Check recent history for fatigue
First scan Content-Learnings/content-index.md (fast, compact) for
category/format/topic repetition in the prior ~2 weeks. For anything it
flags as close, confirm against the actual note (Drafts/, Scheduled/,
Published-Posts/). Avoid assigning two similar posts back-to-back across
week boundaries, not just within the week being planned. If
content-index.md doesn't exist yet or is still thin, fall back to reading
the folders directly, as before.
3. Pick the days, then apply the content-type variety rule
Days: read Content-Learnings/playbook.md first. If it has a
Best-Performing Patterns rule about which specific days perform better,
with real evidence (post ids, ≥3 published posts) attached, use those days
instead of the default below. Otherwise fall back to the starting
heuristic (REQUIREMENTS.md §5): Tue, Thu, Sat, trimmed to
however many slots target_count actually calls for (e.g. target_count:
2 → Tue/Thu). Sunday is never used unless explicitly requested this
invocation.
Content-type variety: unlike the old fixed Mon–Fri table, a specific
content type is no longer bound to a specific day. Instead, across however
many slots this run fills, require at least 2 distinct category/
content_type values among them (checked against the Playbook's evidenced
patterns first, this variety rule otherwise) — don't let 3 slots
accidentally all be the same content type.
4. Match ideas to slots
For each day slot, pick the unassigned candidate idea whose category/
content_type best fits, breaking ties by rank_score, while keeping the
variety rule from step 3 satisfied across the slots filled so far. Do not
force a mismatched or duplicate idea into a slot just to fill it. If no
good fit exists for a slot, leave it unfilled.
5. Commit the assignment
For each idea assigned to a slot:
status: candidate → selectedtarget_week: the ISO week being plannedtarget_date: the specific date
6. Assign a hook formula (filled slots only, REQUIREMENTS.md §32)
For each slot filled by step 5 — never for a slot left empty — read
Content-Learnings/hook-formulas.md fresh (not from memory; it grows over
time via /extract-hook and human promotions). Filter to status:
canonical rows only — a proposed row is not trusted for use here any
more than it is in /write-draft.
- Fit check: prefer a canonical formula whose
engagement_goalsvalue matches the idea's likely engagement goal — the same precedence/write-draftstep 3 uses, but non-binding here: the idea's own goal field if it has one and is set, else aContent-Learnings/playbook.md- evidenced pattern for the idea's category (real post ids attached as evidence), else defaultcomments— and whose mechanic genuinely fits the idea'scategory/content_type. A formula with noengagement_goalsvalue is general-purpose and eligible regardless of goal. - Anti-repetition — two checks, both required:
- This week's own picks: never assign the same
formula_idto two different filled slots within the same plan produced by this run. - Recent history: scan
Content-Learnings/content-index.md'shook_gistcolumn (§15/§25.4) for the last few weeks of published/scheduled posts and avoid a formula whose mechanic was just used — judge this from thehook_gisttext and, where a row's linked note is easy to open, that note's ownhook_formulafield. Ifcontent-index.mddoesn't exist yet or is too thin to judge from, this check is a no-op — don't block on it, same fallback posture as step 2's fatigue check.
- This week's own picks: never assign the same
- No genuine fit: if nothing in the canonical set fits this idea's category/content_type (a goal mismatch alone isn't disqualifying if the mechanic still fits — only a real mechanic mismatch is), leave that slot's hook-formula cell blank in the report and say why in one line — never force a bad fit just to populate the column.
- This assigns a formula suggestion, not the finished hook line —
/write-draftre-readshook-formulas.mditself at drafting time and may pick differently if circumstances have changed by then (a newproposedformula got promoted, playbook evidence shifted, etc.). Nothing here binds/write-draft's own step 5.
7. Assign a posting time (filled slots only, REQUIREMENTS.md §32)
For each filled slot, attach the time exactly as REQUIREMENTS.md §11
already specifies — pure surfacing of existing logic at planning time
instead of leaving it until /schedule-approved; this step adds no new
rule of its own:
- Default: the day's row from §11's table for whichever day step 3 assigned this slot (e.g. Tuesday → 16:00, Thursday → 17:00, Saturday → 09:00).
- Override: if
Content-Learnings/playbook.mdhas aBest-Performing Patternsrule with real evidence (≥3 published posts, post ids attached) for a better day+time+content-type combination covering this slot, use that time instead — the same override step 3 already applies when choosing which days to use; this step applies it again to the time attached to whichever day the slot ends up on.
8. Assign comment targets (filled slots only, REQUIREMENTS.md §32)
For each filled slot, attach 2-3 specific external accounts/posts to proactively comment on that day, in this priority order:
Content-Learnings/comment-targets.md, if it exists and its## Targetstable has rows — a user-maintained list (new in Phase 21, never system-populated). Prefer entries relevant to the slot's topic/category; if several are equally relevant, prefer ones not already used earlier in this same week's plan.- Fallback, only if (1) is missing, absent, or too thin to fill the
slot: accounts/authors/publications that came up as sources or
citations in the research actually backing that slot's idea — i.e.,
a named author or account identifiable in the
sources[]of the Research Notes the idea itself cites from that week's/research-topicruns. Never a generic outlet name with no identifiable account, and never a source from an unrelated topic. - Neither source yields anything: leave the field explicitly empty in the report with a one-line reason (e.g. "no comment-targets.md entries and no research-cited accounts this week") — never invent a plausible-sounding account or post to fill the cell. This is exactly as unacceptable as forcing a hook-formula fit or padding the day count.
9. Report
Render a Mon-Sun calendar table, all seven days, not just the filled
ones: date | day | status | topic | category | format | hook formula |
posting time | comment targets | rank_score.
- Filled rows (
status: filled): every column populated from steps 4-8. - Empty rows (
status: empty): the reason already produced for that gap by step 4 goes directly in the row (e.g. "no Career-category ideas in the pool — run/research-topic Careerthen/generate-ideasbefore this slot can be filled") — shown in the row itself, not buried in separate prose below the table.hook formulaandcomment targetsare blank on every empty row by construction (steps 6 and 8 only ever touch filled slots) — never populate them for a day that didn't clear the bar. - If fewer than
target_countslots ended up filled because the pool genuinely didn't support more distinct, valuable angles, say that plainly in the table and in a closing line — that is the correct outcome per §5, not a shortfall to apologize for.
Hard rules
- Never assign more than
target_count(default 3) posts in one week, and never assign to Sunday unless the user explicitly asks for it this invocation. Saturday is in scope by default. - Never force the count. Filling fewer slots than
target_countbecause nothing else cleared the bar is the correct, expected outcome — never pick a weak or duplicate idea just to reach the target. - Never draft post text here — ideas only get a day, not content.
- Never pick a
status: placeholderidea. - Hook formulas and comment targets are attached only to filled slots (REQUIREMENTS.md §32). Never invent either for an empty day, and never let hook-formula or comment-target scarcity be a reason to fill a slot that otherwise wouldn't clear the quality bar. This rule carries exactly the same weight as the count-padding prohibition above — it is not a softer, secondary concern.
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.

