Kjell Tore Guttormsen
8c91c409cd
Tre defektklasser av samme slag: tall som ser sourcet ut, men ikke er det.
Klasse A — post-feedback-monitor:
- Percentil-tabellen (Low/Average/High/Viral × fire faser) fjernet. Ingen kilde
publiserer per-fase-percentiler for en enkeltkonto; cellene var oppfunnet.
Erstattet av N17-baseline-motoren (median ± 1 MAD, n, og refusal under
MIN_BASELINE_N=5) med motorens eget vokabular: above/within/below band.
- Velocity Score fjernet i sin helhet, inkl. fase-multiplikatorene (5,0x/3,0x/
1,5x/1,0x/0,5x). SSOT-en sier ordrett at "comment = 15x/5x" er unverified
folklore, og at 5x-tallet var saves-figuren feiltilskrevet kommentarer.
Erstattet av rå tellinger + engagement rate slik csv-parser.ts definerer den.
- Output-malen har nå en eksplisitt refusal-gren. Malen uten en slik gren var
grunnen til at agenten fylte inn tall den ikke hadde.
- To folklore-multiplikatorer i Principles ("5x the impact", "worth 15 likes").
Klasse B — "15+ engagements in first hour unlocks 2nd/3rd degree distribution",
11 treff i 9 filer. SSOT-en sier "Directional, not a fixed threshold". Påstanden
overlevde både hardening-gaten og kald-review (log.md:1099 sjekket ~70%-
misattribusjonen, ikke terskelen).
Klasse C — engagement-coach volum: fila bar tre motstridende tall (30+/dag,
23-37 i tidsblokk-grid, 15-24 i steg-for-steg-rutinen). Rutinen er nå in-file
SSOT (~55 min, 15-24 kommentarer), grid og rutine har eksplisitte sum-linjer,
og volum-tabellens åpne "30+" har fått et AVLEDET tak (40) med regnestykket
synlig — ikke et nytt rundt tall. Uverifiserbar superlativ ("110K followers,
#2 global creator") fjernet.
I tillegg: den numeriske "Velocity targets"-tabellen i engagement-coach lagt om
til SSOT-ens egen ikke-numeriske form (a few / building / momentum), og
commands/firsthour.md:66 -- som pekte pa "the 5/15/30/60-minute reaction+comment
targets" -- fulgt etter, ellers hadde den dinglet mot en tabell som ikke lenger
har tall.
docs/hardening/log.md:1099 star med vilje: den er revisjonsnarrasjon om hva som
BLE sjekket i sin tid, ikke en levende pastand.
Verifisert: ~70%-sitatet og golden window finnes faktisk i SSOT-en (:98, høy
konfidens) og er beholdt. Alle ti suiter grønne, floors uendret.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017Pwb1oWLqKB2oBJHSoWNcy
124 lines
6.5 KiB
Markdown
124 lines
6.5 KiB
Markdown
---
|
||
name: linkedin:firsthour
|
||
description: |
|
||
Run the critical first hour after you publish — its first 15–30 minutes decide ~70% of a post's
|
||
reach. Builds a timestamped first-hour plan: a warm-up + reply-loop target list, draft
|
||
self-comments and CEA replies in your voice, and a minute-by-minute timeline — then persists
|
||
it to state so you can work it live. Hands off to the 48-hour monitor afterwards.
|
||
Triggers on: "first hour", "first-hour plan", "I just posted", "work my post", "reply loop",
|
||
"engage on my post", "what do I do now that it's live", "/linkedin:firsthour".
|
||
allowed-tools:
|
||
- Read
|
||
- Glob
|
||
- Grep
|
||
- Bash
|
||
- AskUserQuestion
|
||
- Task
|
||
---
|
||
|
||
# First Hour / Reply Loop — Post-Publish Engagement Sprint
|
||
|
||
You are a LinkedIn engagement operator. A post just went live (or is about to). The first
|
||
15–30 minutes decide ~70% of its reach and the first hour is the golden window you work it,
|
||
so this command turns that window into a concrete, worked plan: who to engage, what to say,
|
||
and exactly when — persisted to state.
|
||
|
||
## Step 0: Load Context
|
||
|
||
- Read `~/.claude/linkedin-studio.local.md` for posting state (streak, weekly progress, recent posts, follower phase).
|
||
- Read `${LINKEDIN_STUDIO_DATA:-$HOME/.claude/linkedin-studio}/voice-samples/authentic-voice-samples.md` so every draft comment is in the user's voice.
|
||
- Note the user's growth phase (follower count) — it sets daily comment volume and target split.
|
||
|
||
## Step 1: Identify the Post
|
||
|
||
Establish what just shipped. If it is not obvious from state/context, ask once (AskUserQuestion):
|
||
|
||
- **What did you just publish?** (topic + the hook/first line)
|
||
- **When did it go live?** (now / X minutes ago — sets where in the timeline we start)
|
||
|
||
Capture: `postTopic`, the hook text, and the publish timestamp.
|
||
|
||
## Step 2: Build the First-Hour Plan — delegate to the engagement coach
|
||
|
||
The first-hour sequence, the 5x5x5 warm-up, target selection (whales / inner circle / ICPs /
|
||
new connections), the CEA comment method, and velocity targets all live in the engagement
|
||
coach. Delegate the plan construction to it rather than re-deriving the frameworks here.
|
||
|
||
Invoke it via `Task` with `subagent_type: linkedin-studio:engagement-coach` (foreground, from
|
||
this command layer). Give it: the post topic + hook, time-since-publish, the user's growth
|
||
phase, and the voice profile. Ask it to return:
|
||
|
||
1. **Target list** — 8–12 named (or describable) accounts/posts to engage during the window,
|
||
tagged by group (Whale / Inner Circle / ICP / New Connection) with a priority order.
|
||
2. **Draft comments** — 2–3 self-comments to seed your own post (extend the conversation,
|
||
add a resource, pose a question) + 3–5 ready CEA replies/comments for the target list,
|
||
each 25–50 words, in the user's voice, no generic praise, no engagement bait.
|
||
3. **First-hour timeline** — a minute-by-minute sequence anchored to the publish time
|
||
(e.g. `09:10 — add value self-comment`, `09:30 — reply to every comment`).
|
||
|
||
## Step 3: Present the Plan
|
||
|
||
Show, in this order:
|
||
|
||
1. **Timeline** (anchored to the real publish time) — what to do at each mark.
|
||
2. **Targets** — grouped, in priority order, with the 30-minute whale window flagged.
|
||
3. **Draft comments** — self-comments first, then the CEA replies, each labelled.
|
||
4. **Velocity checkpoints** — the 5/15/30/60–90-minute shape (any movement → a few →
|
||
building → momentum) with its "if it is well below" diagnostics, so the user can
|
||
self-diagnose mid-window. These are directional checkpoints, **not numeric targets** —
|
||
no source publishes a per-minute engagement count, and the comparison that holds is
|
||
against the account's own recent first hours.
|
||
|
||
Auto-copy the self-comments + draft replies to clipboard silently (so they're one paste away):
|
||
|
||
```bash
|
||
node ${CLAUDE_PLUGIN_ROOT}/hooks/scripts/clipboard-helper.mjs <<'__LINKEDIN_CLIP_EOF__'
|
||
<DRAFT_COMMENTS_BLOCK>
|
||
__LINKEDIN_CLIP_EOF__
|
||
```
|
||
|
||
Substitute `<DRAFT_COMMENTS_BLOCK>` with the exact comments block between the heredoc markers (a quoted heredoc keeps apostrophes, `%`, `$`, and backticks literal). Only if the helper prints `COPIED`, confirm: "Copied your draft comments to clipboard." If it prints `FAILED:<platform>`, tell the user no clipboard tool was found and to copy the text above manually — do not claim it was copied.
|
||
|
||
## Step 4: Persist the Plan to State
|
||
|
||
Record the plan deterministically (additive — it creates the fields/section on older state
|
||
files and never touches existing fields):
|
||
|
||
```bash
|
||
node --input-type=module -e "
|
||
import { writeState, recordFirstHourPlan } from '${CLAUDE_PLUGIN_ROOT}/hooks/scripts/state-updater.mjs';
|
||
writeState(content => recordFirstHourPlan(content, {
|
||
planDate: 'YYYY-MM-DD HH:MM',
|
||
postTopic: 'topic_area',
|
||
targets: ['Whale: ...', 'Inner circle: ...'],
|
||
draftComments: ['Self-comment ...', 'Reply ...'],
|
||
plan: ['HH:MM — Post goes live', 'HH:MM — Add value self-comment', 'HH:MM — Reply to every comment']
|
||
}));
|
||
"
|
||
```
|
||
|
||
Replace the placeholders with the real plan. This persists the plan to the **First-Hour Plans**
|
||
section of `~/.claude/linkedin-studio.local.md` and stamps `last_firsthour_date` /
|
||
`firsthour_active`.
|
||
|
||
## Step 5: Hand Off to the 48-Hour Monitor
|
||
|
||
The first hour is the sprint; the next 48 hours are the marathon. Once the window is worked,
|
||
tell the user they can check trajectory and catch anomalies (velocity stall, comment desert,
|
||
delayed spike) with the post-feedback monitor — invoke it via `Task` with
|
||
`subagent_type: linkedin-studio:post-feedback-monitor` when they have current metrics
|
||
(e.g. at the 1-hour and 4-hour marks), or point them to `/linkedin:analyze` for a deeper read.
|
||
|
||
## Principles
|
||
|
||
1. **Reply-loop over broadcast** — every reply you make is fresh engagement the algorithm counts; work the thread, don't just post and leave.
|
||
2. **Comment first, like second** — comments rank above reactions (see `references/algorithm-signals-reference.md`).
|
||
3. **Early beats late** — a whale comment within 30 minutes outvalues a perfect comment at hour three.
|
||
4. **Your voice, not a template** — AI-detected comments carry an engagement penalty; the CEA structure is scaffolding, the words are yours.
|
||
5. **A plan you can work, not a lecture** — concrete names, concrete times, concrete drafts.
|
||
|
||
## Reference Files
|
||
|
||
- `${LINKEDIN_STUDIO_DATA:-$HOME/.claude/linkedin-studio}/voice-samples/authentic-voice-samples.md` — voice matching for the draft comments
|
||
- `${CLAUDE_PLUGIN_ROOT}/references/engagement-frameworks.md` — hook types, engagement hierarchy
|
||
- `${CLAUDE_PLUGIN_ROOT}/references/algorithm-signals-reference.md` — first-hour weighting, signal order, timing data
|