| name | instagram |
| description | Instagram routing hub for SilphCo / Silph Scope. Use when the user runs /instagram or asks about Instagram Reels, Reels ranking, skip rate, shares, captions, Insights diagnosis, or which Instagram short to ship. Owns the native metrics map (skip/share/like/save/repost/comment) and post-publish primary-action decision tree. Routes to high-retention-insight-reels-instagram for faceless TCG insight Reels create/review/rank. Prefer this hub for any Instagram-first task so specialists stay discoverable under one slash command.
|
| metadata | {"short-description":"Instagram hub — Reels, ranking, Silph Scope"} |
Instagram — routing hub
One entry point for Instagram work. Load this hub first on /instagram, then
open one specialist (default below).
Routing table
| Task | Skill | When |
|---|
| Silph Scope insight Reels (script, storyboard, DistScore rank, ship-gate) | high-retention-insight-reels-instagram | Default for almost all Reels work |
| Generic data-short craft philosophy | data-reels | Density / ship-gate polish only — not distribution rank |
| Social virality research (Gold/Silver/Bronze, multi-platform) | marketing-going-viral | Format research outside Silph mono template |
| Share psychology (STEPPS) | marketing-contagious | Designing send triggers only |
Default path (/instagram with no extra args)
1. Apply native metrics map + (if Insights) post-publish diagnosis from this hub
2. Load high-retention-insight-reels-instagram
3. If local .mp4s: storyboard first (1fps + VO), then DistScore rank
4. If create/rewrite: myth-bust mechanism default; protect first 1.5–3s
5. Report: ship / hold / rewrite-hook (or other primary action) — never equal-weight craft over share
Live account law (Silph Scope, 2026-07)
Observed: Ascended Heroes / Mega Gengar myth-bust ≈ 1.6k views, ~10 DM shares.
Clean comparison / concentration / number-upside Reels: ~70–80% skip, die in test pool.
Until skip rate on new posts is consistently under ~40–45%:
- Gate = first 1–3s (skip). Nothing mid-reel matters if bounce is 60%+.
- Frame-0 / cover hold ≥ ~3.0s — do not fade the thumbnail after ~1s. Hook cover stays until ~3s (sting/main after).
hookHoldMs: 3000 + s1 gapAfterMs in build_audio / scope mix.
- Prefer belief confrontation over clean insight density on frame 0.
- Prefer visual rejection (WRONG / DIDN'T CRASH / red penalty chip) over calm +green badges.
- Portable complete rule at the end (no dangling “because—”).
- Innovate inside confrontation energy — do not clone Gengar frame-for-frame; do not revert to soft data layouts as the open.
Native metrics map (always apply)
Instagram Reels Insights order under “What impacts your views” (rates listed by importance to reach). Use this table on every create, rank, and postmortem — do not invent a different priority order.
Skip vs retention (diagnostic split)
| Metric | Window | Diagnoses |
|---|
| Skip rate | Leave in first ~3s | Hook only — frame 0 claim, text, motion, pattern interrupt |
| Retention / avg % watched / curve | After 3s through end | Story + pacing — mid beats, dead air, payoff timing, length |
A low skip with a steep mid drop is not a hook problem. A high skip with fine mid craft is only a hook problem — rewrite open, do not polish the cascade first.
Reach-impact order (native UI)
| # | Rate | What it measures | Healthy (industry / Silph) | Kill / weak |
|---|
| 1 | Skip rate | % scroll away in first ~3s | ~30–40% skip; Silph gate until <~40–45% | ≥50–60% — dies in test pool |
| 2 | Share rate | Sends / shares per reach (DMs count hardest) | Explicit send trigger + portable conflict | Near-zero sends on non-follower push |
| 3 | Like rate | Likes per reach | Secondary social proof | Chase likes alone = wrong optimization |
| 4 | Save rate | Saves per reach | Final portable rule / reference numbers | No rule, pure opinion dump |
| 5 | Repost rate | Public reposts per reach | Same drivers as share when content is “for the timeline” | Ignore unless Insights shows it as a lever |
| 6 | Comment rate | Comments per reach | Lowest of the six for reach; use for conversation, not rank #1 | Do not over-optimize CTAs that hurt retention |
Stack that sits under those rates
| Signal | Role |
|---|
| Total watch time + replays | Once past skip gate — aggregate seconds and loops (Mosseri-class) |
| Sends per reach (DM) | Strongest distribution share form; often cited ~3–5× a like for non-followers |
| Completion / watch-through | Support metric on short Reels; not a free pass if skip is 70% |
| Early velocity / test pool | Weak first-pool skip+share → throttled; strong → expanded |
Ship hierarchy for Silph (do not reverse): skip gate → share/DM → save → watch/replays/completion → likes → repost/comment.
DistScore (specialist) = Hook + Tension + Share + Save. Map: Hook↔skip, Share↔share/DM, Save↔save; Tension feeds retention after the gate.
Post-publish diagnosis playbook
Run after any published Reel with Insights (or user-reported rates). Output one primary action — do not stack “fix everything.”
Inputs to collect
Skip rate · share/sends (count or rate) · save rate · likes · avg % watched or retention notes · reposts/comments if shown · template used (myth-bust / concentration / H2H / other).
Decision tree (first matching row wins)
| If | Then primary action | Do not |
|---|
| Skip ≥ ~50–60% (or Silph soft-open pattern ~70%+) | rewrite-hook — frame 0 belief attack, visual rejection, muted-legible claim in ≤1.5–3s | Polish mid-reel density, length, or end CTA first |
| Skip OK (≲40–45%) but sends/shares ≈ 0 and weak travel | add-share-trigger — name the wrong belief + “send to anyone who thinks…”; make the rule quotable | Assume “needs more data density” |
| Skip OK, some watch, saves flat, no memorable rule | add-save-payload — one portable rule + hold final numbers as screenshot bait | Add more facts without a single rule |
| Skip OK, steep drop mid-reel (retention cliff after ~3s) | cut-mid — kill true dead air, logo stings, multi-second wipes; retime payoffs every ~3–5s | Rewrite a working hook |
| Skip OK, mid OK, completion soft only because length | tighten-or-split — one insight per Reel; trim after the rule | Default “always shorter” dogma if cascade needs ~18–22s and skip already clears |
| All rates weak on a soft comparison / calm +% open | kill-template for growth bets — switch to myth-bust / confrontation until skip clears account gate | Ship another concentration-as-growth open |
| Skip + sends strong (Gengar-class: travel + DMs) | clone-mechanism not frames — keep confrontation energy + share trigger; new data/topic | Frame-clone art; or revert to soft opens because “variety” |
| Likes high, skip or share bad | Ignore vanity — fix skip or share per rows above | Optimize for more likes |
Silph template read
| Pattern | Read |
|---|
| Myth-bust / belief confrontation clears skip + earns DMs | Default growth template; innovate inside it |
| Concentration / clean H2H / number-upside dies at 70%+ skip | Hold as polish/save craft only until account skip gate clears |
| Same template, skip OK once then skip high next | Hook/topic heat failed — rewrite-hook, not abandon mechanism |
Required postmortem shape
## Insights read
- Skip: X% → hook: pass|fail
- Share/sends: … → travel: pass|fail
- Save: … · avg watched / cliffs: …
- Template: …
## Primary action
rewrite-hook | add-share-trigger | add-save-payload | cut-mid | tighten-or-split | kill-template | clone-mechanism
## Next ship constraint
- One sentence (e.g. "Frame 0 must reject a named belief; no calm +green open.")
When not to use this hub
| Task | Use instead |
|---|
| Offers / landing / email | marketing hub |
| Product app UI | product-design |
| Data pipelines | data |
Done criteria
