Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    onchain flow attribution

    1

    Accurately attribute blockchain flows by separating internal transfers from external entity demand across multiple chains.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    onchain flow attribution

    onchain flow attribution

    Example session with this skill installed

    Analyze the transfer records in input.json for Entity_Alpha on Ethereum and Polygon. Compare against the Q3 baseline and identify flow anomalies or attribution gaps.

    • Read your context and instructions
    • Compiled the onchain flow attribution

    I have processed the transfer records for Entity_Alpha. The analysis normalized 1,240 events across Ethereum and Polygon, successfully filtering internal reshuffling from net external flows.

    On-Chain Flow Attribution Report

    MetricEthereum (ETH)Polygon (POL)Combined (USD Eq.)
    Total Transfers850390-
    Internal Transfers610120-
    Net External Inflow145.502,100.00$412,400.00
    Net External Outflow94.00550.00$268,100.00
    Attribution Gap5.2%1.8%-
    Anomaly Score+12% vs Baseline-4% vs BaselineModerate Alert

    Observations

    Internal Shuffle: A high volume of transfers (610) on Ethereum were identified as internal movements between Entity_Alpha's hot and cold storage, correctly excluded from external flow totals.

    Anomalous Inflow: Ethereum inflows are 12% above the Q3 baseline, primarily driven by three large-value transfers from unlinked CEX deposit addresses.

    Attribution Gaps: 44 transfers remained unclassified due to lack of evidence-backed labels; these have been excluded from the reviewed entity totals to maintain data integrity.

    Data References

    • Source: examples/input.json
    • Helper Version: 1.0.4
    • Baseline Period: 2023-Q7-01 to 2023-09-30

    Next steps

    1. Review the 44 unclassified transfers in report.json to verify if new wallet clusters should be defined.
    2. Cross-reference the $412k inflow with market news timestamped at 14:00 UTC to check for coordinated selling intent.
    3. Run the bridge-pairing script to reconcile the Polygon outflow with Ethereum mint events.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Attributing blockchain transfer data manually leads to double-counting and false positives. Most tools fail to distinguish between internal entity reshuffling and genuine external demand, skewing risk and volume metrics.

    What it does

    • Normalizes multi-chain transfer logs using chain-qualified address keys and event-level deduplication.
    • Separates internal transfers within the same reviewed entity from external inflows and outflows.
    • Aggregates flows per chain and token while maintaining strict unit scale isolation.
    • Identifies anomalies by comparing current flow snapshots against historical baselines and market context.
    • Flags attribution gaps and uncertain labels for manual review instead of forcing inaccurate classifications.

    Frameworks & tools

    Python 3 for analysis scripts, JSON for data input/output, and chain-specific transaction logs.

    Why this beats prompting it yourself

    General LLMs often hallucinate entity relationships or fail to handle the precision required for log index deduplication. This skill enforces a strict normalization and attribution logic that prevents common accounting errors like merging cross-chain address strings or misinterpreting bridge legs.

    Use cases

    • Auditing exchange inflow vs. internal wallet management.
    • Monitoring for anomalous whale movements against a historical baseline.
    • Separating bridge mint/burn semantics from market-driven transfers.

    Known limitations

    Does not include node indexers, automated wallet clustering, or cross-chain bridge matching. Unknown attributions are flagged but excluded from reviewed totals.

    No automated wallet clustering or node connectors. Requires pre-labeled entity data. Does not include built-in statistical anomaly thresholds or sanctions classifiers.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Requires

    Python

    Permissions

    Terminal / Shell
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    It deduplicates supplied transfer events by chain, transaction hash, and log index, and excludes transfers between reviewed addresses belonging to the same entity. It does not determine whether an external flow represents demand.

    The skill is designed as a modular Python-based analysis tool that accepts JSON inputs, making it compatible with any AI agent capable of executing Python scripts or handling structured data logs.

    The package includes the instructions, local Python analysis helper, JSON input contract, synthetic fixtures, example report, runtime notes, smoke-test script, and logo. Entity labels and transaction data must be supplied separately.

    No, this skill focuses on the attribution logic itself; you will need to provide your own transaction logs from a node provider or indexer as the data source.

    By using chain-qualified address keys, the skill treats addresses on different networks as distinct entities, preventing the common error of merging identical strings across incompatible chains.

    The transfer is flagged as uncertain and excluded from reviewed entity totals until its attribution is resolved. The tool does not guess wallet ownership or bridge relationships.

    New in Trading & Investing