gate-info-tokenonchain
General Rules
⚠️ STOP — You MUST read and strictly follow the shared runtime rules before proceeding.
Do NOT select or call any tool until all rules are read. These rules have the highest priority.
→ Read
gate-runtime-rules.md
→ Also read
info-news-runtime-rules.md for
gate-info /
gate-news-specific rules (tool degradation, report standards, security, routing degradation, per-skill version checks when
is present, and legacy wrapper routing).
- Only call MCP tools explicitly listed in this skill. Tools not documented here must NOT be called, even if they
exist in the MCP server.
- Legacy / routing mode: when Step 0 emits , use only the MCP tools listed in this file. When Step 0 emits , do not call those MCP tools; delegate to the mapped primary skill per Step 0.
Token On-Chain Analysis Skill (current version: no Smart Money). For token-level holder distribution, activity, and large transfers, call on-chain + basic coin info tools in parallel, then aggregate into a structured report.
Trigger Scenarios: User asks about token on-chain data, holder distribution, on-chain activity, large transfers, on-chain chip analysis, etc.
Per-skill updates: This directory includes
and may include
.
ClawHub packages uploaded to the marketplace often omit
(upload-page restriction);
GitHub / Bitbucket source trees keep both. Policy:
info-news-runtime-rules.md §1. The
Trigger update steps below apply in addition to
gate-runtime-rules.md; when scripts are present, use this flow for version checks before execution.
Update check — user visibility: Technical failures during version check (missing script, sandbox, network, non-zero exit, no parseable
line) must
not be explained to the user; continue with Execution per
info-news-runtime-rules.md. Only
successful /
outcomes may be summarized (including
/ strict
exit 3, which is still a
success path that requires user confirmation before
).
Do not auto-download
from the network.
Static reference if
is missing: canonical scripts live in
gate/gate-skills under
(same
as frontmatter).
Step 0 — Wrapper routing probe
This legacy skill is a compatibility alias for the primary CLI skill
.
Before Trigger update, MCP tool selection, or any legacy Execution Workflow,
run a deterministic shell probe:
bash
PRIMARY_SKILL="gate-info-web3"
HAS_PRIMARY=0
for root in \
"$HOME/.cursor/skills" \
"$HOME/.codex/skills" \
"$HOME/.openclaw/skills" \
"$HOME/.agents/skills" \
"$HOME/.gemini/antigravity/skills"
do
if [ -f "$root/$PRIMARY_SKILL/SKILL.md" ]; then
HAS_PRIMARY=1
break
fi
done
if command -v gate-cli >/dev/null 2>&1 && [ "$HAS_PRIMARY" = "1" ]; then
echo "__ROUTE_CLI__"
else
echo "__FALLBACK__"
fi
Interpretation:
- → stop here. Do not run Trigger update or the legacy MCP
sections below. Follow
https://github.com/gate/gate-skills/blob/master/skills/gate-info-web3/SKILL.md
- → continue with Trigger update and the legacy MCP workflow in
this file.
Trigger update (with Execution)
Default (PD / blocking): On each trigger, run
first
without GATE_SKILL_UPDATE_MODE=auto
. The script
only compares local vs remote; if
, it
does not overwrite files.
STOP — ask the user whether to update
before any
,
, or skill Execution. In sandboxes (e.g. Cursor), use
full / all permissions when later running
. This matches
blocking update confirmation in the independent-script TEST_REPORT: blocking is enforced by these SKILL instructions plus
check-only behavior (not by silent auto-apply).
Strict check (Agent / IDE — recommended on ): Use
GATE_SKILL_CHECK_STRICT=1
. When
,
exits 3 and prints
— it stops
only that shell step.
Hard rule: after you ask the user,
end this assistant turn; do
not call
,
, or MCP in the
same message.
applies without a prior user step —
do not use
in chat for the blocking flow; use
→ user confirms → .
Step 1 — Check
Install root: /
read
. Typical
:
$HOME/.cursor/skills/<name>/
,
$HOME/.codex/skills/<name>/
,
$HOME/.openclaw/skills/<name>/
,
$HOME/.agents/skills/<name>/
,
$HOME/.gemini/antigravity/skills/<name>/
. Single-arg: the script resolves
in that order when
exists; otherwise
is
. Workspace-only trees need two-arg
/
with explicit
.
Agent selection rules (run
at most one of the fenced examples below; choose by OS and files under
):
- macOS / Linux / WSL / Git Bash: If exists → use only the Bash example.
- Windows (native PowerShell): If exists → use only the PowerShell example.
- Windows without (typical ClawHub install): If only exists → try WSL or Git Bash () using the same command shape as the Bash example. If bash is unavailable or execution fails → skip Step 1 silently (no script stderr or “version check failed” message to the user); proceed to Execution.
- Neither script exists: Skip Step 1 silently; proceed to Execution. Do not invent or run / .
Bash (blocking — no auto; example Cursor):
bash
GATE_SKILL_CHECK_STRICT=1 bash "$HOME/.cursor/skills/gate-info-tokenonchain/scripts/update-skill.sh" check "gate-info-tokenonchain"
PowerShell:
powershell
$env:GATE_SKILL_CHECK_STRICT = '1'
powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\.cursor\skills\gate-info-tokenonchain\scripts\update-skill.ps1" check "gate-info-tokenonchain"
Result semantics: = no action.
= remote newer;
with blocking flow, do not apply until the user agrees.
= could not compare — proceed with current version per
info-news-runtime-rules.md;
do not surface technical check failure details to the user.
Agent parse (stdout): GATE_SKILL_UPDATE_AGENT_ACTION=…
.
BLOCK_UNTIL_USER_CONFIRMS_UPDATE
→ Step 2 before Execution.
→ no block from the check script.
Step 2 — Confirm or Reject (blocking)
Runtime: Use the
same shell family for Step 2 as for Step 1 (Bash vs PowerShell). If Step 1 was
skipped, do
not run
or
.
-
STOP — do NOT proceed to Execution yet.
-
Inform the user (e.g. newer version available; summarize if helpful).
-
Wait for the user’s reply — blocking step.
Hard rule (Cursor / Agent): When
reports
, or
BLOCK_UNTIL_USER_CONFIRMS_UPDATE
, or strict
,
end this turn after asking.
Only in the
user’s next message run
(if they agree) or
(if they decline). Do
not chain
in the same turn as
for this flow.
- User agrees → run with from strict stdout when required, then Execution.
- User declines → , then Execution on the current install.
Two-step gate (strict ): /
(without
GATE_SKILL_UPDATE_MODE=auto
)
fail until
matches
. User decline →
.
bash
GATE_SKILL_CONFIRM_TOKEN="<paste from check stdout>" bash "$HOME/.cursor/skills/gate-info-tokenonchain/scripts/update-skill.sh" apply "gate-info-tokenonchain"
bash
bash "$HOME/.cursor/skills/gate-info-tokenonchain/scripts/update-skill.sh" revoke-pending "gate-info-tokenonchain"
powershell
$env:GATE_SKILL_CONFIRM_TOKEN = '<paste from check stdout>'
powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\.cursor\skills\gate-info-tokenonchain\scripts\update-skill.ps1" apply "gate-info-tokenonchain"
powershell
powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\.cursor\skills\gate-info-tokenonchain\scripts\update-skill.ps1" revoke-pending "gate-info-tokenonchain"
If Step 1 was not strict (no pending token):
without
is allowed.
If or : no update step; proceed to Execution.
Optional — GATE_SKILL_UPDATE_MODE=auto
For
CI / unattended automation only: setting
GATE_SKILL_UPDATE_MODE=auto
on
makes the script
apply immediately when the remote is newer —
no user confirmation and
incompatible with
blocking update confirmation tests. Do
not use
on
when reproducing the blocking PD flow.
Parameters
- name: Frontmatter above; must match on gate-skills.
- Invoke: Use
$SKILL_ROOT/scripts/update-skill.sh
~/.openclaw/skills/<name>
~/.gemini/antigravity/skills/<name>
- ClawHub vs full tree: Installs without may copy it from gate/gate-skills under (manual only; agents must not auto-download).
Do not dump raw script logs into the user-facing reply except when debugging. On
exit 3 (strict), do not run Execution until Step 2 is resolved. On
or
failure, still run Execution when appropriate per runtime rules.
Known Limitations
- Smart Money:
info_onchain_get_smart_money
info_onchain_get_token_onchain
- Entity profiling:
info_onchain_get_entity_profile
- On-chain data coverage depends on upstream chain support (e.g., BlockInfo).
MCP Dependencies
Legacy path only — this section applies when Step 0 emitted
.
Required MCP Servers
| MCP Server | Status |
|---|
| Gate-Info | ✅ Required |
MCP Tools Used
Query Operations (Read-only)
- info_onchain_get_token_onchain
- info_coin_get_coin_info
Authentication
- API Key Required: No
- Credentials Source: None; this skill uses read-only Gate Info / Gate News MCP access only.
Installation Check
- Required: Gate-Info
- Install: Use the local Gate MCP installation flow for the current host IDE before continuing.
- Continue only after the required Gate MCP server is available in the current environment.
Routing Rules
Legacy path only — when Step 0 emitted
, routing is delegated to
.
| User Intent | Keywords | Action |
|---|
| Token holder distribution | "ETH holders" "BTC holding distribution" "top holders" | Execute with |
| On-chain activity | "on-chain activity" "active addresses" "transaction count" | Execute with |
| Large transfers | "large transfers" "whale movements" "unusual transfers" | Execute with |
| Full on-chain overview | "on-chain analysis for SOL" "ETH on-chain data" | Execute with holders,activity,transfers
|
| Smart Money (not yet supported) | "smart money buying" | Inform user; run available scopes only |
| Specific address query | "track this address 0x..." | Route to |
| Coin fundamentals | "analyze SOL" | Route to |
| Whale entity tracking | "what is Jump Trading doing" | Route to if available, else inform |
Execution Workflow
Legacy path only — this section applies when Step 0 emitted
.
Step 0: Multi-Dimension Intent Check
- Token-level on-chain → this Skill.
- Specific address (not token) → .
- Fundamentals + technicals + news together → (if available).
Step 1: Intent Recognition & Parameter Extraction
- (required): Token ticker (e.g., BTC, ETH, SOL)
- (optional): e.g., eth, sol, bsc
- : one or more of , , ( — not available)
- (optional): default 24h for transfers/activity
Step 2: Call MCP Tools in Parallel
| Step | MCP Tool | Parameters | Retrieved Data | Parallel |
|---|
| 1a | info_onchain_get_token_onchain
| symbol, chain, scope, time_range
| Holder / activity / transfer data per scope | Yes |
| 1b | | query={symbol}, scope="basic"
| Basic coin context | Yes |
Step 3: LLM Aggregation
- Contextualize on-chain data with coin info
- Identify patterns and anomalies
- Avoid speculative price predictions
Report Template
Legacy path only — this section applies when Step 0 emitted
.
markdown
## {symbol} On-Chain Analysis
> Generated: {timestamp} | Chain: {chain or "All supported chains"}
> Note: Smart Money analysis not yet available in this version.
### Token Overview
|--------|-------|
| Token | {symbol} ({name}) |
| Market Cap | ${market_cap} |
| Circulating Supply | {circulating_supply} |
### Holder Distribution (if scope includes holders)
{Tables + LLM concentration assessment}
### On-Chain Activity (if scope includes activity)
{Metrics + LLM trend assessment}
### Large Transfers (if scope includes transfers)
{Table + LLM flow assessment}
### On-Chain Health Score
{Dimensions scored /10 + overall}
### Key Insights
{2–3 data-driven bullets}
> On-chain data does not predict future prices. This does not constitute investment advice.
Decision Logic
| Condition | Assessment |
|---|
| Top 10 holder concentration > 70% | High concentration risk |
| Top 10 holder concentration < 30% | Well-distributed holder base |
| Active addresses declining > 20% WoW | Declining network activity |
| Active addresses growing > 30% WoW | Strong activity growth |
| Large transfers to exchange addresses | Potential sell pressure |
| Large transfers from exchange addresses | Potential accumulation |
| User asks about Smart Money | State not available; offer holders/activity/transfers |
Error Handling
| Error Type | Handling |
|---|
| Token not found on-chain | Check via ; suggest symbol/chain |
info_onchain_get_token_onchain
| Show coin info only; note on-chain unavailable |
| fails | Show on-chain data without market context |
| Scope returns empty | Skip section; note no data for scope |
| Chain not supported | List supported chains; ask user |
| Both Tools fail | Return error; suggest retry later |
| User requests | Inform not available; offer other scopes |
Cross-Skill Routing
| User Follow-up Intent | Route To |
|---|
| "Analyze this coin" | |
| "Track this address" | |
| "Is this token safe?" | |
| "Technical analysis?" | |
| "Any news?" | |
| "What does the community think?" | |
| "DeFi data for this?" | |
Safety Rules
- No fabricated on-chain data: Only report MCP-returned data.
- Address privacy: Shorten addresses (e.g., ); do not doxx.
- No trading signals: Informational only; not buy/sell advice.
- Exchange labels: Best-effort; may be mislabeled.
- Data lag: Note indexing delays where relevant.
- Smart Money: Clearly state unavailability rather than approximating.
- Age & eligibility: Intended for users aged 18 or above with full civil capacity in their jurisdiction.
- Data flow: The host agent processes user prompts; this skill directs read-only Gate-Info MCP tools listed above. The LLM summarizes tool output. This skill does not invoke additional third-party data services.