agentsclimarketplace

Defi data fetcher

Skill auralshin/agent-skills/skills/defi-data-fetcher

DEFI-focused skill pack for agentic assistants

Install
npx -y skills add auralshin/agent-skills --skill defi-data-fetcher

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use this skill when users need DeFi market/protocol data (TVL, APY, volume, fees, prices) gathered, normalized, reconciled across sources, and returned with explicit freshness/confidence metadata.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.4 KB, 764 tokens by cl100k_base, as published. Nobody here has run it

DeFi Data Fetcher

Purpose

Collect DeFi metrics from prioritized sources, normalize them, reconcile cross-source conflicts, and return a source-attributed dataset with freshness and confidence labels.

Use this skill when

  • The user asks for current or historical DeFi metrics (TVL, APY, volume, fees, revenue, token prices).
  • The user wants protocol/token comparisons across chains.
  • The user needs a clean dataset before risk or strategy analysis.

Do not use this skill when

  • The task is transaction signing or broadcasting.
  • The task is pure protocol economic risk scoring (use defi-risk-evaluator).

External dependency profile

  • Dependency level: High for live/current metrics.
  • Primary sources: protocol-native APIs/subgraphs and official analytics.
  • Secondary sources: DeFiLlama and market data aggregators.
  • Validation/backfill: direct RPC reads.
  • Offline fallback: supports normalization/reconciliation/reporting on user-provided snapshots only.

Workflow

  1. Clarify query scope:
    • Protocols/tokens/chains
    • Time window (latest, 24h, 7d, custom)
    • Required metrics
  2. Build source plan with references/source-priority.md.
  3. Fetch using ordered providers and keep retrieval timestamps.
  4. Normalize fields/units via references/metric-definitions.md.
  5. Apply freshness policy from references/freshness-sla.md.
  6. Reconcile conflicts (median + spread analysis) and assign confidence.
  7. If live fetch is unavailable, switch to references/offline-fallback.md mode and state limits.
  8. Return required schema.

Data quality rules

  • Always separate apy_base and apy_reward.
  • Percentages are decimal internally (0.12 = 12%).
  • All timestamps must be UTC ISO-8601.
  • Never hide source disagreement; show spread and confidence.
  • Explicitly flag stale or partial coverage.

Required output format

{
  "query_scope": {
    "protocols": ["string"],
    "chains": ["string"],
    "time_window": "string",
    "requested_metrics": ["string"]
  },
  "fetch_mode": "live|offline_snapshot",
  "source_plan": {
    "primary": ["string"],
    "secondary": ["string"],
    "validation": ["string"]
  },
  "metrics": [
    {
      "metric": "tvl_usd|apy_base|apy_reward|volume_24h_usd|fees_24h_usd|revenue_24h_usd|price_usd",
      "entity": "protocol_or_token",
      "chain": "string",
      "value": 0,
      "as_of": "ISO-8601",
      "freshness_status": "fresh|stale|unknown",
      "confidence": "high|medium|low",
      "spread_pct": 0,
      "sources": ["string"]
    }
  ],
  "reconciliation_notes": ["string"],
  "quality_flags": ["string"],
  "summary": "2-4 sentence summary"
}

Bundled resources

  • references/metric-definitions.md: Canonical metric semantics.
  • references/source-priority.md: Source ranking and failover policy.
  • references/freshness-sla.md: Metric-specific freshness thresholds.
  • references/offline-fallback.md: Behavior when live providers are unavailable.
  • scripts/normalize_metrics.py: Deterministic normalization + optional reconciliation mode.

Use scripts/normalize_metrics.py --reconcile when you have multiple rows per metric/entity/chain and need consistent confidence/spread outputs.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.