gmgn-skills/docs/workflow-early-project-screening.md

Early Project Screening — Workflow

Use this workflow to rapidly screen newly launched tokens from launchpads and identify whether any are worth a closer look, before committing to a full token research deep dive.

Use this workflow when:

• "early project screening" / "are any new tokens worth accumulating?"
• "screen the latest launched tokens for me"
• "which new tokens have smart money entering early?"
• "any new tokens on pump.fun worth watching?"
• User wants to filter new launchpad tokens for quality signals before buying

---

Step 1 — Fetch Newly Launched Tokens

# Tokens just created, still on bonding curve
gmgn-cli market trenches --chain <chain> --type new_creation

# Tokens near bonding curve completion (about to graduate to DEX)
gmgn-cli market trenches --chain <chain> --type near_completion

# Tokens that have already graduated to open market
gmgn-cli market trenches --chain <chain> --type completed

Which type to use:

• new_creation — highest upside potential, highest risk. Many will fail.
• near_completion — approaching graduation, momentum building. Tighter window.
• completed — already trading on DEX, more liquidity but earlier gains may be gone.

From the results, note each token's address, symbol, smart_degen_count, renowned_count, volume, swaps, and rug_ratio.

Tip — use filter flags to pre-screen at fetch time:

# Fetch with safe baseline filter (server-side)
gmgn-cli market trenches --chain <chain> \
  --type new_creation --type near_completion \
  --filter-preset safe --sort-by smart_degen_count

# Strict: safe + require smart money + min 24h volume $1k
gmgn-cli market trenches --chain <chain> \
  --type new_creation --type near_completion \
  --filter-preset strict --sort-by smart_degen_count

# Custom: manual range filters (all sent server-side)
gmgn-cli market trenches --chain <chain> \
  --type new_creation \
  --max-rug-ratio 0.3 --max-bundler-rate 0.3 --max-insider-ratio 0.3 \
  --min-smart-degen-count 1 --min-volume-24h 1000

Using --filter-preset safe (or strict) tells the server to pre-filter results before returning — equivalent to Steps 2's "Discard immediately" criteria, applied before the response is sent.

---

Step 2 — First-Pass Filter (In-Response Scan)

If you used --filter-preset safe or --filter-preset strict in Step 1, the rug_ratio, bundler_rate, and insider_ratio checks below are already applied server-side. Verify the remaining signals manually.

Before running any CLI commands per token, apply a quick in-response filter on the trenches results:

Discard immediately if:

• rug_ratio > 0.3
• is_wash_trading = true
• bundler_rate > 0.3
• rat_trader_amount_rate > 0.3
• Zero smart_degen_count AND zero renowned_count AND volume < $10k

Keep for deeper screening if any of:

• smart_degen_count ≥ 1 (smart money has entered)
• renowned_count ≥ 1 (KOL has entered)
• bluechip_owner_percentage > 0 (quality wallet base)
• volume is strong relative to token age

Select up to 5 tokens that pass this filter.

---

Step 3 — Security Check (Per Token)

For each shortlisted token:

gmgn-cli token security --chain <chain> --address <token_address>

Hard stops — discard token immediately if:

Field	Hard Stop
is_honeypot	"yes" (BSC/Base)
renounced_mint (SOL)	false
renounced_freeze_account (SOL)	false
rug_ratio	> 0.3
sell_tax	> 0.10
top_10_holder_rate	> 0.6

Proceed with tokens that pass all hard stops.

---

Step 4 — Smart Money Early Entry Check

# Who's already in? (smart money holders)
gmgn-cli token holders --chain <chain> --address <token_address> \
  --tag smart_degen --order-by buy_volume_cur --direction desc --limit 10

# Top traders — any known wallets?
gmgn-cli token traders --chain <chain> --address <token_address> \
  --order-by profit --direction desc --limit 10

Strong signal:

• Smart money wallets entered early (check buy_30m / buy_1h counts on the holder — or cross-reference last_active_timestamp being recent)
• Multiple distinct smart money wallets (not one large wallet — that's concentration risk)
• Top traders show profit (token has already rewarded early holders, positive momentum)

Weak signal:

• Only one smart money wallet in
• Smart money wallet entered but profit is negative (they're underwater)

---

Step 5 — Token Info Spot Check

gmgn-cli token info --chain <chain> --address <token_address>

Check:

• Social presence: link.twitter_username, link.telegram, link.website — at least one should exist
• holder_count — growing is a positive sign
• wallet_tags_stat.smart_wallets — confirms smart money count
• cto_flag — if 1, community has taken over, dev is gone (neutral to positive, evaluate context)
• creator_token_status — creator_close means dev has sold (mixed: reduces dump risk, but also less team commitment for very new tokens)

---

Screening Output

Present results as a table, then a per-token verdict:

Early Project Screening — {chain} / {type}
Screened: {N} tokens from trenches → {M} passed filter

# | Symbol | Address (short) | Smart Degens | Rug Risk | Security | Verdict
1 | ...     | ...             | {N} wallets  | {X}      | ✅/⚠️/🚫  | Watch / Small position / Skip
...

─── Top Pick ───────────────────────────────
{SYMBOL}: Smart money in early, security clean, social present
→ Suggested action: Small exploratory position / Watch for 1h / Skip

Verdict scale:

• 🟢 Small position — clean security + smart money early entry + social presence
• 🟡 Watch — some positive signals but missing key indicators; monitor for 30–60 min
• 🔴 Skip — any hard stop triggered, or no smart money interest at all

---

Follow-Up Actions

For any token rated 🟢:

• Run full due diligence: workflow-token-research.md
• Check risk warnings before sizing up: workflow-risk-warning.md
• Execute swap if satisfied: use gmgn-swap skill

---

Related Workflows

• workflow-token-research.md — full 5-step token analysis
• workflow-risk-warning.md — ongoing risk monitoring for held positions
• workflow-market-opportunities.md — trending token discovery (graduated tokens with volume)