li-reply
Skill2 starsChanged 43 days agoMITView on GitHub
Draft replies to every comment under the user's own LinkedIn post, triaged into lead, substance, peer, support, and noise, and answered in that order. Use when the user pastes their comment thread, asks "help me reply", or wants to work through responses on a published post. Writes drafts only; never replies on their behalf.
8 other projects use a version of this.
Adopted by
Most recently taken up by
- furkanbilalakyuz/linkedin-agent-skill-trtheir own version, first seen Nov 8, 2026
- mkmlabs-ai/linkedin-agenttheir own version, first seen Nov 8, 2026
- mubaid/dsh-linkedin-agenttheir own version, first seen Oct 25, 2026
- robbiehaynes/job-searchtheir own version, first seen Oct 20, 2026
- postzen-dev/linkedin-agent-skilltheir own version, first seen Oct 15, 2026
- avi691/linkedin-agent-skilltheir own version, first seen Oct 8, 2026
Other versions8 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 8
- Jakeschincariol/linkedin-agent-skill★ 514same name
- avi691/linkedin-agent-skill★ 5same name
- mkmlabs-ai/linkedin-agent★ 4same name
- nessalazne/linkedin-agent-skill★ 3same name
- postzen-dev/linkedin-agent-skill★ 3same name
- furkanbilalakyuz/linkedin-agent-skill-tr★ 0same name
- mubaid/dsh-linkedin-agent★ 0same name
- robbiehaynes/job-search★ 0same 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 repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- li-auditSkill
- li-carouselSkill
- li-commentSkill
- li-dmSkill
- li-humanSkill
- li-inboxSkill
- li-planSkill
- li-postSkill
- li-profileSkill
- li-repurposeSkill
- li-voiceSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 44
Peers: treat as a conversation between equals.
- Distinctive: line 48
Keep every reply under 60 words.
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
li-reply
The thread under a good post is where relationships start. This skill triages the whole thread and drafts every reply in the user's voice, in priority order. The user posts the replies themselves.
Before you write
- Load the voice profile (
li-voiceif missing). - Get the thread: the user pastes the comments, or describes them. You need the commenter's name and the comment text for each.
Triage
Sort every comment into exactly one bucket:
| Bucket | What it looks like | Priority |
| Lead | Asks about the work, the offer, the service; "how do you do X for clients" | 1 - answer first |
| Substance | Disagrees, adds data, asks a real question | 2 |
| Peer | Industry peer adding their take or tagging someone relevant | 3 |
| Support | "This is great", congrats, emoji-only | 4 - one line, warm, move on |
| Noise | Spam, pitch-slaps, off-topic, hostile | 5 - do not draft; say "ignore" or "delete" |
Show the triage as a compact list before any drafts, so the user can correct your read. A misclassified lead is expensive.
Drafting rules
- Leads: answer the actual question, then one soft next step ("happy to walk you through it - DM me"). Never pitch in the open thread beyond that.
- Substance: engage the point, not the person. If they are right, say so plainly; it builds more authority than winning.
- Peers: treat as a conversation between equals. Add something, don't just agree.
- Support: one genuine line. Never "Thanks! 🙏" on repeat; vary it or say nothing.
- Keep every reply under 60 words. The thread belongs to the commenters; the user is the host, not the keynote.
- Match reply depth to comment depth: a three-word comment gets a sentence, not a paragraph.
The loop
1. Triage list. Commenter, bucket, one-line reason.
2. Drafts in priority order. Each in its own fenced block, labeled with the commenter name and bucket.
3. Humanize. Invoke the li-human skill across all drafts before showing
them. Report the lowest score of the batch.
4. Never publish. The user copies and posts each reply. On "yes", log to
~/.agents/linkedin/log.md with the date and post identified by its first
line.
Example
TRIAGE
- Maya Chen / LEAD: asks how we price the audit she just read about
- Tom R. / SUBSTANCE: disagrees with the 25% figure, cites his own data
- Priya S. / SUPPORT: "needed this today"
- growth_hacker_99 / NOISE: pitch for lead-gen services -> ignoreDoes 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.