Agent skill
post-tge-growth-system
Audit a Web3 protocol's growth system in the first months after token launch.
Filed under Positioning and messaging.
From gogrowth-co/web3-protocol-gtm-skills · 3 skills · 0 · pushed 2026-08-03
What it does when it runs
Audit a Web3 protocol's growth system in the first months after token launch. Use when a founder says "we launched and growth stalled," "TVL is bleeding," "holders are leaving," or asks what to fix first post-TGE. Separates real growth failure from market beta, maps where the funnel leaks, and produces a prioritized 90-day fix list.
Read from the skill and the 0 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.
- Keys and connectors you must supply
- None found.
- Hosts it reaches
- mangabeira.net
- Tool permissions it declares
- No
allowed-toolsin the frontmatter. It only issues instructions, so there is nothing to bound. - Actions present in the files
- None. Instructions only.
Install it
View source on GitHub ↗git clone --depth 1 --filter=blob:none --sparse https://github.com/gogrowth-co/web3-protocol-gtm-skills.git /tmp/web3-protocol-gtm-skills git -C /tmp/web3-protocol-gtm-skills sparse-checkout set "post-tge-growth-system" mkdir -p ~/.claude/skills/post-tge-growth-system cp -R "/tmp/web3-protocol-gtm-skills/post-tge-growth-system/." ~/.claude/skills/post-tge-growth-system/
Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗
The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.
The skill
Source on GitHub ↗Reproduced in full from gogrowth-co/web3-protocol-gtm-skills/blob/d92fb9ed94c3f89c44720511f62eed1391dcdc4d/post-tge-growth-system/SKILL.md, which is licensed MIT (repository). 867 words, 18 headings.
Post-TGE Growth System
Diagnose why growth stalled after token launch, and what to fix first. The output is a leak map with severity flags and a 90-day priority list — not a pep talk and not a list of 40 tactics.
This is the on-chain and systems layer of the 6-pillar Web3 Growth Audit methodology (dApp + Community pillars plus the attribution stack), adapted for self-serve use.
When to use
- Token launched in the last ~6 months and price/TVL/holders are declining
- Wallet connects happen but transactions don't follow
- The team can't tell which marketing spend produced which on-chain result
- Deciding what to fix first with a small team and limited runway
Inputs to collect first
- Token ticker + chain, launch date
- Where the numbers live: DefiLlama/Dune/CoinGecko links if they have them
- The core on-chain action a user should take (stake, deposit, mint, trade)
- What marketing they're currently running
- Their attribution setup, if any (GA4, Cookie3, UTM discipline, none)
Workflow
Step 1 — Beta check: is this actually a growth failure?
Before diagnosing anything, separate protocol failure from market beta. Web3 tokens are heavily correlated to BTC.
- Price: token return over the window minus BTC return over the same window
- TVL: protocol TVL trend minus sector-wide TVL trend (DefiLlama)
A token down 40% while BTC is down 35% is market movement, not a marketing failure — and "fixing marketing" won't fix it. A token meaningfully underperforming its benchmark has a real narrative or distribution problem.
State the verdict explicitly before continuing. If the protocol merely tracked the market down, say so — that finding alone prevents a quarter of wasted panic spend.
Step 2 — Funnel walk
Trace the actual user path: discovery → website → wallet connect → first transaction → repeat transaction. For each step, what does the data show (or what data is missing)?
The most common leak: connects without first transactions. The dApp has no activation sequence — nothing guides a connected wallet to the core action. Mark each step ✅ / ⚠ / ❌ with evidence.
Step 3 — Activation loop inventory
List every mechanism that turns a new user into a repeat user: incentives, streaks, notifications, email sequences, quests, community roles tied to on-chain actions. For each: does it bridge to the core on-chain action, or does it reward activity that never touches the chain?
- At least one loop that reliably produces repeat core actions: ✅
- Loops exist but reward off-chain activity (Discord messages, X raids): ⚠ High
- No loops — growth is 100% acquisition, 0% retention: ❌ Critical
Step 4 — Attribution stack check
- UTM discipline: are campaign links tagged, consistently?
- Wallet-connect capture: does the UTM survive to the connect event?
- On-chain join: can any acquisition source be tied to any transaction?
If the answer to the third is no: every budget decision is a guess. Mark ❌ Critical. This is the single most common post-TGE infrastructure gap.
Step 5 — Owned audience check
Same check as the distribution scan, because post-TGE it becomes acute: when incentives taper, can the protocol reach its users without paying an algorithm? Email list ✅ / signup-only ⚠ / nothing ❌ High.
Step 6 — Synthesis
One paragraph answering: where does this protocol have presence without a system? Tie the leaks together into a single diagnosis. The pattern to look for:
- Channels exist, system doesn't
- Impressions accumulate, users don't
- The on-chain funnel leaks before it starts
- The attribution gap collapses budget defensibility
- Owned audience is zero
Output format
# Post-TGE Growth System — [Protocol]
## Beta check
[Underperforming benchmark, or tracking the market? Numbers + verdict.]
## Leak map
| Funnel step | Status | Severity | Evidence |
|---|---|---|---|
| Discovery → site | ... | ... | ... |
| Site → connect | ... | ... | ... |
| Connect → first TX | ... | ... | ... |
| First TX → repeat | ... | ... | ... |
| Attribution | ... | ... | ... |
| Owned audience | ... | ... | ... |
## Synthesis
[The one-paragraph diagnosis: presence without a system, and where.]
## 90-day priority list
[Exactly three fixes, ordered. For each: what, why it's first/second/third,
and the observable signal that proves it worked. Fixing the biggest leak
beats optimizing any single channel.]
Rules
- Never skip the beta check. Diagnosing a market-wide drawdown as a marketing failure is the most expensive mistake this skill exists to prevent.
- Every severity flag needs evidence: a number, a screenshot-able state, or an explicit "no data exists — which is the finding."
- Three fixes, not ten. A post-TGE team in decline has no capacity for ten.
Going deeper
This skill runs on public data and the team's own numbers. The full Web3 Growth Audit adds Dune cohort retention curves, connect→TX funnel queries, wallet-cohort analysis by acquisition source, and a full 90-day roadmap with a recorded walkthrough, delivered async with a fixed scope and price:
Other skills for the same job
Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.
- growth-engine by ericosiu · 3,449
- x-longform-post by ericosiu · 3,449
- growth-strategy by OpenClaudia · 664
- growth-loops by AIDevGTM · 252
- building-gtm-system by GTM-Strategist · 245
- post-meeting-follow-up-automation by louisblythe · 136
- growth-strategy by manojbajaj95 · 92
- blog-post-writer by markster-public · 61
Need help setting it up?
This page tells you what post-tge-growth-system does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.
Book a call →The directory stays free. There is nothing gated behind this.