pre-launch-update
Automates the maintenance of stablecoin launch information and project summaries.
Install
mkdir -p .claude/skills/pre-launch-update && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/18674" && unzip -o skill.zip -d .claude/skills/pre-launch-update && rm skill.zipInstalls to .claude/skills/pre-launch-update
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Use when asked to update pre-launch stablecoin data, add milestones, refresh launch phases, or check for promotions. Runs weekly to keep the upcoming stablecoins module current.Key capabilities
- →Update pre-launch stablecoin data
- →Add milestones to stablecoin projects
- →Refresh launch phases for stablecoins
- →Check for promotions of stablecoins
- →Update AI summaries for stablecoins
How it works
The skill runs weekly to update pre-launch stablecoin data by researching current information from various sources. It updates milestones, launch phases, and AI summaries based on material changes.
Inputs & outputs
When to use pre-launch-update
- →Updating stablecoin status
- →Adding launch milestones
- →Refreshing project data
About this skill
Pre-Launch Module Update
Weekly full-lifecycle update of all pre-launch stablecoins tracked by Pharos. Updates milestones, launch phases, expected dates, featured content, and AI summaries.
Scope of Updates
Valid enum values for launchPhase, milestone type, and featuredContent.type live in shared/types/core.ts as LAUNCH_PHASE_VALUES, LAUNCH_MILESTONE_TYPE_VALUES, and FEATURED_CONTENT_TYPE_VALUES. Read that file for the authoritative list — this document may drift.
Extended reference (edge cases, history, examples): read ./reference.md when needed.
Data Locations
Pre-launch coins live in shared/data/stablecoins/coins/*.json with status: "pre-launch", and are sourced in generated runtime output via shared/data/stablecoins/coins.generated.json and canonical-order.json.
AI summaries: data/ai-summaries.json
Canonical types & enums: shared/types/core.ts.
Material Change Definition
Update the AI summary in data/ai-summaries.json only when any of these occur:
launchPhaseadvancesexpectedLaunchDateshifts by ≥1 quarter (forward or backward)- New named regulator, custodian, or major partner (bank, exchange, chain infra, payments network)
- Audit completed by a reputable firm
- Regulatory approval, charter, or license granted or denied (e.g., BitLicense, EMI, MiCA, state charter)
- Public announcement of a confirmed mainnet deployment date
Skip cosmetic updates: single promotional tweets, blog restyles, minor PR mentions, secondary partnerships. When in doubt, do not rewrite.
Process
Step 0 — Staleness radar
Rank coins by staleness before researching so attention lands on the most-outdated entries.
- For each coin, find the most recent
milestones[].date(or the coin's last edit viagit log -1 --format="%cs" -- shared/data/stablecoins/coins/*.jsonif milestones are empty) - Present a sorted list (oldest first) with days-since-last-milestone
- Do deep research (Twitter fetch, full web search, browser fallback) on the top 5 stalest coins; lighter-touch checks on the rest
Any coin with no milestones, or with a latest milestone older than 8 weeks, is an automatic deep-research candidate and also a stall-flag candidate (the issuer may have quietly delayed or abandoned the project).
Also read the deferred-items ledger at agents/pre-launch-deferred.md (create it on first run). It records prior runs' dead-ends — preview-only listings, lapsed dates with no sourced replacement, unresolved identity matches — each with a checked-on date and a re-check trigger. For ledger items, do a cheap re-verification against the trigger (usually one API call) instead of a full deep-research pass, unless the trigger fires or the entry is older than 4 weeks.
Step 1 — Read current state
- Read all
shared/data/stablecoins/coins/*.jsonentries wherestatusis"pre-launch"and list every coin with its currentlaunchPhase,expectedLaunchDate,launchPhaseDetail, milestone count, andfeaturedContentcount - Read
data/ai-summaries.jsonto see which pre-launch coins have summaries and when they were last updated - Present a summary table to the user: coin name, phase, expected date, milestones count, last summary update
Step 2 — Research each coin
For each pre-launch coin, run these searches in parallel where possible. Prioritize depth on coins flagged stalest in Step 0.
- Twitter/X: Check the coin's Twitter handle (from
links[]entries labeled "Twitter"). Look for tweets about: launch dates, milestones, partnerships, regulatory approvals, testnet/mainnet activity, delays- Fallback order:
WebFetch https://x.com/{handle}first. On 403, only escalate to the browser fallback (claude-in-chrome or Playwright in Claude Code;agent-browserin Codex) if the coin is a deep-research candidate from Step 0 or if another source already indicates fresh signal. Do not burn browser sessions on every coin. - Look for tweets from the last 7-14 days
- Fallback order:
- Official website:
WebFetchthe coin's website (fromlinks[]entries labeled "Website") for press releases, blog posts, status updates - Web search:
WebSearchfor"{coin name}" stablecointo find: news articles, regulatory filings, partnership announcements - Link liveness: HEAD-check (or GET-and-check-status) every entry in
links[]plus the top 3featuredContent[].urlvalues. Flag any non-2xx responses in the findings report so dead references can be fixed or replaced in Step 4. - DefiLlama/CoinGecko: Check if the coin has appeared on either platform (search by name/symbol). If found, evaluate against promotion criteria in Step 5 — do not flag based on listing existence alone
- DefiLlama:
WebFetchhttps://stablecoins.llama.fi/stablecoinsand search response for the coin name/symbol - CoinGecko:
WebFetchhttps://api.coingecko.com/api/v3/search?query={symbol}— then fetch the detail endpoint to readmarket_data.market_cap.usdandcirculating_supply
- DefiLlama:
Step 3 — Present findings
For each coin with updates, present:
## {Name} ({Symbol}) — {current phase}
### Changes found
- **launchPhase**: {old} → {new} (source: {URL})
- **expectedLaunchDate**: {old} → {new} (source: {URL})
- **New milestones**:
- {date} | {type} | {title} (source: {URL})
- **New featuredContent**:
- {type} | {title} | {URL}
### Promotion candidate?
{Yes/No — explain why}
### No changes
{List coins where nothing new was found}
Important: Flag any conflicts between sources. If unsure about a value, say so explicitly.
Step 4 — Apply changes (after per-coin user approval)
Approval is per coin, not bulk. For each coin with proposed changes, wait for explicit user confirmation (e.g., "yes / approved / go") before editing that coin's entry.
- Edit the coin's entry in the appropriate JSON file using the
Edittool - When updating
expectedLaunchDate, you must follow the Date History Protocol (below) first — no exceptions, even for minor shifts inside the same year - Add new milestones to the
milestones[]array, sorted oldest-first by date - Add new featured content to
featuredContent[] - Update AI summaries in
data/ai-summaries.jsononly for coins that meet the Material Change Definition (followwrite-ai-summariesvoice guidelines) - Verify:
npm run check:stablecoin-data— validates the zod schema for the tracked registry (including pre-launch entries). Required.- If the approved changes will be published, use
pharos-release-runnerafter the focused checks. GitHub's requiredPR gateis authoritative; the heavy local merge gate is optional rehearsal work. npm run buildalone is not sufficient — it does not enforce the pre-launch schema.
Step 5 — Flag promotions (with explicit evidence)
A coin is ready-to-promote only if all three of the following hold.
Before recommending promotion, also run stablecoin-runtime-price-marketcap-gate and record the accepted active runtime path. A promoted asset must have both a fetchable current price and a fetchable market-cap / circulating-supply path.
- Listed externally with real supply:
- DefiLlama: entry on
https://stablecoins.llama.fi/stablecoinswith non-zerocirculating, OR - CoinGecko: entry reachable via
https://api.coingecko.com/api/v3/search?query={symbol}whose detail endpoint returns non-zeromarket_data.market_cap.usdand non-zerocirculating_supply
- DefiLlama: entry on
- Identity match:
- If
geckoIdis already set on the pre-launch entry, the discovered CoinGecko id must match. - If not set, cross-verify symbol plus issuer domain (match
links[]Website host against the issuer link on the listing page). Symbol-only matches are unsafe (e.g., generic tickers collide).
- If
- Not a preview listing: market cap, circulating supply, and 24h volume are all zero → it's a preview, not a live listing. Flag for tracking, not promotion.
Report format:
- Ready:
**{Name} ready for promotion** — {source}, circulating=${X}, geckoId={id}, identity match confirmed - Preview only:
**{Name} has a preview listing, not ready** — CoinGecko entry exists but circulating/market-cap are zero; keep tracking - Ambiguous:
**{Name} possible match, unverified** — {reason}; recommend manual confirmation before next run
Do NOT execute the promotion (removing status: "pre-launch", adding llamaId, normalizing across chains, moving the entry out of shared/data/stablecoins/coins/*.json by status change, etc.) — that's a manual, coordinated process that requires schema changes beyond the scope of this skill.
After reporting, update agents/pre-launch-deferred.md: one line per unresolved outcome ({coin} — {finding} — checked {YYYY-MM-DD} — re-check when {trigger}). Refresh the checked date on re-verified items and delete entries that resolved this run.
Date History Protocol
Mandatory. Every change to expectedLaunchDate — including minor shifts inside the same year (e.g., 2026-Q2 → 2026-Q3) — must append a dateHistory entry before the new value is written.
When expectedLaunchDate changes:
- Read the current
expectedLaunchDatevalue - Append
{ "date": "{current value}", "setOn": "{today YYYY-MM-DD}" }to thedateHistory[]array (create the array if it doesn't exist) - Then update
expectedLaunchDateto the new value
Do not backfill dateHistory from guessed prior values. Only the first shift observed under this protocol gets recorded — if a coin has no dateHistory today, leave it empty until an actual live shift is observed.
Milestone Guidelines
Each milestone entry:
{
"date": "2026-03-22",
"type": "milestone",
"title": "Private testnet goes live",
"description": "Optional longer description",
"sourceUrl": "https://example.com/announcement"
}
- date: Always
YYYY-MM-DDformat - type: Must be one of the values in
LAUNCH_MILESTONE_TYPE_VALUES. Read `share
Content truncated.
When not to use it
- →When executing promotions from pre-launch to active status
- →When removing existing milestones
- →When changing `flags` (backing, pegCurrency, governance) or `id`, `name`, `symbol`
Limitations
- →Does not execute promotions from pre-launch to active
- →Does not remove existing milestones or change canonical identifiers
- →AI summary updates are only triggered by material changes
How it compares
This skill automates the weekly update and research process for pre-launch stablecoins, ensuring data consistency and flagging staleness, unlike manual data entry and research.
Compared to similar skills
pre-launch-update side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| pre-launch-update (this skill) | 0 | 15d | No flags | Intermediate |
| quant-analyst | 103 | 2mo | No flags | Advanced |
| stock-analyzer | 71 | 1mo | Review | Beginner |
| creating-financial-models | 36 | 8mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by TokenBrice
View all by TokenBrice →You might also like
quant-analyst
zenobi-us
Expert quantitative analyst specializing in financial modeling, algorithmic trading, and risk analytics. Masters statistical methods, derivatives pricing, and high-frequency trading with focus on mathematical rigor, performance optimization, and profitable strategy development.
stock-analyzer
FrancyJGLisboa
Provides comprehensive technical analysis for stocks and ETFs using RSI, MACD, Bollinger Bands, and other indicators. Activates when user requests stock analysis, technical indicators, trading signals, or market data for specific ticker symbols.
creating-financial-models
anthropics
This skill provides an advanced financial modeling suite with DCF analysis, sensitivity testing, Monte Carlo simulations, and scenario planning for investment decisions
reconciliation
anthropics
Reconcile accounts by comparing GL balances to subledgers, bank statements, or third-party data. Use when performing bank reconciliations, GL-to-subledger recs, intercompany reconciliations, or identifying and categorizing reconciling items.
analyzing-financial-statements
anthropics
This skill calculates key financial ratios and metrics from financial statement data for investment analysis
us-stock-analysis
tradermonty
Comprehensive US stock analysis including fundamental analysis (financial metrics, business quality, valuation), technical analysis (indicators, chart patterns, support/resistance), stock comparisons, and investment report generation. Use when user requests analysis of US stock tickers (e.g., "analyze AAPL", "compare TSLA vs NVDA", "give me a report on Microsoft"), evaluation of financial metrics, technical chart analysis, or investment recommendations for American stocks.