Cypher Factory Launchpad

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This is a coherent Ethereum trading plugin, but it combines real wallet-signing authority with autonomous transaction features and registry/dependency disclosure gaps that deserve review before use.

Use read-only mode unless you intentionally want the agent to trade. If enabling writes, use a dedicated wallet with limited funds, verify the resolved SDK package, set low TxGate limits and explicit confirmations, use dry-run for strategies, and monitor or stop any running autonomous strategy.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal engine telemetry is currently stale for this artifact.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
High
What this means

A model-initiated tool call or started strategy could make irreversible trades or other on-chain transactions up to the configured TxGate limits.

Why it was flagged

The tool surface can spend funds on Ethereum Mainnet and includes autonomous trading behavior, while human confirmation is configurable rather than mandatory.

Skill content
With a wallet configured, the plugin can perform real on-chain transactions ... copy-trading, and autonomous strategies ... Confirm Above ETH | No | Confirmation ceiling; 0 disables the feature
Recommendation

Use a dedicated low-balance wallet, set conservative per-transaction and daily TxGate limits, enable a nonzero confirmation threshold where supported, and test strategies with dry-run mode first.

#
ASI03: Identity and Privilege Abuse
High
What this means

Users may not receive the expected credential warning before granting a wallet private key or keystore password that can authorize real Ethereum transactions.

Why it was flagged

The supplied registry credential summary does not surface the private-key or keystore credential path, even though the package documentation and manifest describe those credentials as enabling wallet signing for write operations.

Skill content
Env var declarations: none ... Primary credential: none
Recommendation

The registry metadata should declare the optional primary credentials clearly; users should only provide credentials for a segregated wallet they are willing to let the agent trade from.

#
ASI09: Human-Agent Trust Exploitation
Medium
What this means

A user could underestimate what will run locally or how much authority the plugin has when installed.

Why it was flagged

This user-facing install summary conflicts with the provided package artifacts that declare executable Node plugin code. For a wallet-signing plugin, calling it instruction-only understates the runtime impact.

Skill content
Install specifications
No install spec — this is an instruction-only skill.
Recommendation

Treat this as an executable plugin, not an instruction-only skill, and verify the manifest, package entry point, and installed dependency versions before enabling wallet write mode.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

Core transaction-signing behavior depends on an external package whose exact resolved code is not established by the provided package metadata.

Why it was flagged

The plugin imports the SDK for launchpad agent logic, wallet handling, transaction gating, and strategies, but the dependency is not pinned to a concrete reviewed version in this artifact.

Skill content
"dependencies": {
    "@cypherindustries/factory-sdk": "workspace:^"
  }
Recommendation

Install only from a trusted release channel, verify the resolved SDK package/version, and prefer lockfile or pinned-release provenance for production wallets.

#
ASI10: Rogue Agents
Low
What this means

After first use, the plugin may keep polling the token registry and appending audit logs while the host process remains active.

Why it was flagged

The plugin discloses startup initialization and recurring activity after first tool use. This is purpose-aligned but means it can continue polling/logging beyond a single prompt.

Skill content
"activation": { "onStartup": true ... "backgroundActivityOnFirstToolInvocation": ["TokenRegistry 60s polling loop ...", "Audit-log JSONL writer at ~/.cypher/audit/ ..."] }
Recommendation

Disable or unload the plugin when not needed, and stop any running strategy tools explicitly.

#
ASI06: Memory and Context Poisoning
Low
What this means

Local files may retain transaction history, strategy state, and an encrypted wallet artifact across restarts.

Why it was flagged

The plugin persists audit logs, state, and an encrypted wallet file. This is disclosed and aligned with the trading purpose, but persisted strategy/state data can affect later sessions.

Skill content
"writePaths": ["~/.cypher/audit/", "~/.cypher-factory/state/", "~/.cypher-factory/agent-wallet.enc"]
Recommendation

Protect the host account and filesystem, review retained audit/state files, and remove the generated wallet/state if you stop using the plugin.