wjs-promoting-skills

Runs automatically every day at 4:00 AM: select a

wjs-*

skill → generate today's promotion angle → post to X → draft community posts. X posts are published directly, while community posts are only drafted to outbox/ for manual review.

Core Principle

Claude is the brain, bash is the skeleton. All decisions of this skill (which skill to promote today, how to write the copy, what angle to use) are handled via headless calls to

claude -p

— bash only takes care of scheduling, passing context, logging, and preventing duplicate posts. This means you don't need to modify code when adding new skills, new channels, or changing styles; you only need to adjust prompt templates.

Only one channel supports direct publishing: X. Reddit / HN / Discord don't have reliable APIs for "automatic posting as a user" (either requiring OAuth + anti-spam review, or no API at all). So this skill only drafts community posts to
outbox/<date>/
, waiting for manual copy-paste.

Idempotent + Throttling. Post at most 1 X tweet per day. The same skill won't be promoted again within 7 days. If SKILL.md hasn't changed since the last promotion and there have been no updates in the past 7 days, skip the day. This prevents the bot from being overly noisy.

When This Skill Fires

User says "帮我推广这些 skills" / "设置每天自动发推广" / "研究一下别人怎么推广 skill 的"
User explicitly calls
```
/wjs-promoting-skills
```
User asks "今天准备推哪个 skill" / "outbox 里有什么"

If the user only wants to post once (without cron), use

/publish-skill <name>

directly — that's the manual single-post version. The value of this skill is daily automation.

File Layout

~/.claude/skills/wjs-promoting-skills/
├── SKILL.md                                # This file
├── setup.sh                                # One-time: install launchd plist, start 4 AM schedule
├── uninstall.sh                            # One-time: uninstall launchd plist
├── daily.sh                                # ★ 4 AM entry point: run the full workflow
├── list-skills.sh                          # List all current wjs-* skills
├── pick-next-skill.sh                      # Select today's skill according to rotation rules
├── research-marketplaces.sh                # Research how top skills on openclaw / clawhub / SkillsMP write copy (run once a month)
├── make-plan.sh                            # Generate a marketing plan for a skill (30-day angle rotation)
├── com.jianshuo.wjs-promoting-skills.plist.template
├── prompts/
│   ├── research-marketplaces.md            # Research prompt
│   ├── make-plan.md                        # Marketing plan prompt
│   ├── daily-post.md                       # Today's X post prompt
│   └── community-drafts.md                 # Reddit / HN / Discord draft prompt
├── state/                                  # State directory (.gitignored, won't be pushed to GitHub by publish hook)
│   ├── .gitignore
│   ├── README.md
│   ├── research.md                         # Output of the latest marketplace research
│   ├── plans/<skill>.md                    # 30-day angle rotation plan for each skill
│   └── history.jsonl                       # Append one line after each run: date / skill / tweet_id / status
└── outbox/                                 # Community post drafts (copy-paste after manual review)
    └── YYYY-MM-DD/
        ├── x-posted.txt                    # Archived text of the actually posted X tweet
        ├── reddit-r-ClaudeAI.md
        ├── hn-show.md
        ├── discord-anthropic.md
        └── wechat-followup.md              # (Optional) Angle reference for future WeChat official account articles

Setup (one-time)

bash

~/.claude/skills/wjs-promoting-skills/setup.sh

It will do three things:

Check prerequisites:
```
claude
```
CLI,
```
xurl
```
(and
```
xurl whoami
```
returns the user),
```
jq
```
Run
```
research-marketplaces.sh
```
once to generate initial
```
state/research.md
```
(this is the only time it will actually research online, it will auto-refresh once a month next time)

Render

com.jianshuo.wjs-promoting-skills.plist.template

into a real plist and place it in

~/Library/LaunchAgents/

, then run

launchctl bootstrap

After completion, it will trigger automatically at 04:00 every day without any manual operation.

To stop it:

~/.claude/skills/wjs-promoting-skills/uninstall.sh

Daily Flow (4 AM — what

daily.sh

does)

04:00 →
  Step 1: list-skills.sh        → List of all current wjs-* skills + their last promotion time
  Step 2: pick-next-skill.sh    → Select today's skill according to rules (see §Rotation Rules)
  Step 3: Check skip conditions → Promoted within 7 days / SKILL.md unchanged / Today's content already in outbox → exit 0 directly
  Step 4: make-plan.sh <skill>  → If the skill has no plan or the plan is older than 30 days, regenerate it
  Step 5: Run daily-post.md via claude -p
            └── Input: SKILL.md + plan + research.md + last 30 days of history.jsonl
            └── Output: An X text ≤ 280 characters (including repo URL)
  Step 6: xurl -X POST -d '{"text": "..."}' /2/tweets   ← Publish directly
  Step 7: Run community-drafts.md via claude -p
            └── Output: outbox/<date>/reddit-r-ClaudeAI.md / hn-show.md / discord-anthropic.md
  Step 8: Append one line to history.jsonl, and archive the posted X tweet to outbox/<date>/x-posted.txt

After running daily flow every Sunday, additionally:

```
research-marketplaces.sh
```
refreshes
```
state/research.md
```
(check changes in copywriting of top skills on marketplaces that week)

After running daily flow on the 1st of every month, additionally:

Re-run
```
make-plan.sh
```
for all skills with plans older than 30 days (rotate angles)

Rotation Rules (pick-next-skill.sh)

Sorted by priority from highest to lowest:

Skills never promoted before — Prioritize these to get new skills exposure early
Skills not promoted in the last 7 days — Among these, select those with recently modified SKILL.md (indicating the user just iterated on it)
Others — Select the skill that was promoted the longest ago among all skills

If all skills have been promoted within 7 days → Skip the day (avoid noise). If all skills have never been promoted → Select the first one in alphabetical order.

X Post Format (daily-post.md output)

Constraints:

≤ 280 characters (X counts URLs as 23 characters, reserve 25)
First line: skill name + one-sentence value proposition
Middle: 2–3 specific capabilities / differences from the latest update
Second-to-last line: repo URL on a separate line (X will automatically render a preview card)
At most 1 hashtag (
```
#ClaudeCode
```
or
```
#ClaudeSkills
```
)
Prohibited: Marketing jargon, rocket emojis,
```
AI-powered
```
,
```
game-changer
```
, exaggerated words
Required: Match the user's tone (refer to the style of wjs-* skills in README.md: practical, specific, no hype)

Angle rotation (no repetition when promoting the same skill multiple times):

Angle A: Specific problem it solves (details like "75-minute 4K multi-camera 60GB re-encoding sync doubles disk usage")
Angle B: Counterintuitive design decisions (like "sidecar over re-encode")
Angle C: Workflow integration (what results can be achieved with other skills)
Angle D: Differences from the latest update (if SKILL.md has changed since last promotion, focus on new features from the diff)

make-plan.sh

will arrange these four angles into a 30-day rotation, with a pre-written hook sentence for each.

Community Drafts (community-drafts.md output)

Not published directly. Saved to

outbox/<date>/

, copy-paste after manual review:

reddit-r-ClaudeAI.md
: Long-form format, start with a hook, body covers scenarios + code snippets + links, end with an open question (Reddit prefers discussions over ads). Selfpost only, never post pure links
hn-show.md
: Show HN format, title
```
Show HN: <skill> – <one line value>
```
, body < 600 words, note open source license + repo URL, don't sell on HN
discord-anthropic.md
: Short post, 2–3 sentences + link, suitable for #show-and-tell channels
wechat-followup.md
(Optional): If the skill is suitable for a detailed WeChat official account article, provide a 200-word outline + several angles for direct use when the user writes the article later

If the user is too busy to review on the day, it won't repost — the posts will stay in the outbox, and new drafts (for another skill) will be generated the next day.

Manual Override

For testing / debugging — any step can be run individually:

bash

# See who will be promoted tomorrow
~/.claude/skills/wjs-promoting-skills/pick-next-skill.sh

# Force generate a plan for a specific skill
~/.claude/skills/wjs-promoting-skills/make-plan.sh wjs-transcribing-audio

# Run the entire today's flow without posting to X (dry-run)
DRY_RUN=1 ~/.claude/skills/wjs-promoting-skills/daily.sh

# Force promote a specific skill (bypass rotation)
SKILL=wjs-segmenting-video ~/.claude/skills/wjs-promoting-skills/daily.sh

# Refresh marketplace research
~/.claude/skills/wjs-promoting-skills/research-marketplaces.sh

# See what's been promoted in the last 7 days
tail -7 ~/.claude/skills/wjs-promoting-skills/state/history.jsonl | jq .

Prerequisites

```
claude
```
CLI is in
```
$PATH
```
, and
```
claude -p "hello"
```
runs successfully (headless mode)
```
xurl
```
is installed,
```
xurl whoami
```
returns
```
jianshuo
```
```
jq
```
is installed (
```
brew install jq
```
)
Public repositories: Each
```
wjs-*
```
skill has a GitHub repo, and
```
git config --get remote.origin.url
```
can retrieve the URL. This repository
```
~/code/claude-skills
```
already meets this requirement
macOS (uses
```
launchd
```
for scheduling; the Linux equivalent is systemd timer, but this skill doesn't implement it)

Anti-Patterns (Absolutely Not Do)

Not Do	Reason
Post multiple X tweets a day	7000 followers don't need a cluttered timeline. 1 post per day, quality > quantity
Auto-post to Reddit / HN	Anti-spam reviews are extremely strict, auto-posting = account ban risk. Draft only
Use templates (emojis + exaggerated words)	The user's tone is "practical, specific, no hype". Template-style copy pollutes the brand
Promote the same skill within 7 days	Even with different angles, wrong frequency will cause followers to unfollow
"Post for the sake of posting" — post even if SKILL.md hasn't changed in 7 days	Don't post if there's nothing new. Silence is better than noise
Call `claude` without `--bare`	We need deterministic output, no hooks / auto-memory / CLAUDE.md interference
Push state/ to GitHub	history.jsonl contains metadata of X posts (like tweet_id), which is not public content

What This Skill Is NOT

Not "let AI run your account for you" — every post is based on your written SKILL.md, AI only reframes it from different angles
Not a "cross-platform cross-poster" — only X supports direct publishing, others are drafts
Not "fully autonomous" — you can stop it anytime with uninstall.sh, or use
```
DRY_RUN=1
```
to see what will be posted next time
Not a replacement for
```
/publish-skill
```
— that's the manual single-post version, this is the daily automated version, they can coexist

Done When

setup.sh

runs successfully,

launchctl list | grep wjs-promoting-skills

shows it

```
state/research.md
```
exists and is > 1KB (indicates research has run)
```
DRY_RUN=1 daily.sh
```
runs completely without posting to X
4 markdown drafts exist in
```
outbox/<today>/
```
A new line is added to
```
history.jsonl
```
on the day of actual posting

Auto-publish: This skill is automatically synced to github.com/jianshuo/claude-skills via
~/.claude/skills-publish-hook.sh
(auto commit + push after each edit).
state/
and
outbox/
are excluded via the directory's
.gitignore
+ subdirectory
.gitignore
, and won't be pushed along with SKILL.md.

wjs-promoting-skills

NPX Install

Tags

SKILL.md Content (Chinese)

wjs-promoting-skills

Core Principle

When This Skill Fires

File Layout

Setup (one-time)

Daily Flow (4 AM — what
`daily.sh`
does)

Rotation Rules (pick-next-skill.sh)

X Post Format (daily-post.md output)

Community Drafts (community-drafts.md output)

Manual Override

Prerequisites

Anti-Patterns (Absolutely Not Do)

What This Skill Is NOT

Done When