Moltpho
v0.1.5Shop autonomously on Amazon via Moltpho - search products, manage credit, and purchase items using mUSD on Base mainnet
⭐ 1· 2.1k·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The stated purpose (autonomous shopping on Amazon using mUSD) matches the code and API endpoints (catalog search, quotes, x402 signing, orders). Storing agent credentials and performing on-chain payment signing is consistent with that purpose. Minor inconsistency: SKILL.md references an optional environment override (MOLTPHO_CREDENTIALS_PATH) but the skill metadata lists no required env vars.
Instruction Scope
Runtime instructions ask the agent to register, read/write a credentials file in the user's home/appdata, auto-open a browser, and run a proactive_monitoring loop that passively watches conversations for 'need' signals and may autonomously create quotes and place orders. Passive monitoring + automatic purchase flows are scope-expansive and can lead to unexpected charges if owner confirmation or portal settings are not strictly enforced. SKILL.md also relies on reading/writing local credential files and optionally pulling an 'openclaw_instance_id' without documenting its source.
Install Mechanism
This is listed as an instruction-only skill but includes Python code and a requirements.txt (requests>=2.28.0) with no install spec. That mismatch means the runtime environment may lack declared dependencies or installation steps, causing failures or unexpected behavior. Because there is no install hook, it's unclear how/when dependencies will be installed.
Credentials
The skill does not declare required environment variables yet the SKILL.md documents an override variable (MOLTPHO_CREDENTIALS_PATH). The skill stores API key secrets and wallet addresses to disk (~/.config/moltpho/credentials.json or %APPDATA% path) and will save an api_key_secret shown once by the service — storing such secrets locally is expected for this purpose but is sensitive and should be highlighted to owners. No other unrelated credentials are requested, which is appropriate.
Persistence & Privilege
always:false (good), but the skill is allowed to be invoked autonomously (platform default) and explicitly implements proactive/autonomous purchasing. That combination increases blast radius (ability to create real orders and spend credit) even without 'always: true'. There is no evidence it tries to modify other skills or global settings.
What to consider before installing
Key things to consider before installing: (1) This skill can autonomously detect 'need' signals in conversations and place real orders—if you enable it, set confirmation_required in the portal or disable proactive/autonomous purchasing to avoid unexpected charges. (2) It will create and store API credentials and secrets on disk (~/.config/moltpho/credentials.json or %APPDATA%); treat those like sensitive keys and verify the service (api.moltpho.com / portal.moltpho.com) is legitimate before saving secrets. (3) The package includes Python code and requirements.txt but no install instructions — confirm how dependencies (requests) will be provided or run it in a sandbox/isolated environment. (4) There's a minor mismatch: the SKILL.md references MOLTPHO_CREDENTIALS_PATH but the skill metadata lists no env vars — expect possible configuration surprises. (5) The skill source/owner is unknown; verify the publisher and, if possible, audit the code (especially bootstrap, credential storage, and the proactive purchase logic) and restrict per-order/daily caps in the portal before enabling autonomous behavior.Like a lobster shell, security has layers — review code before you run it.
latestvk9712b0hr7km0x0c8qvnbzmpns80njyj
License
MIT-0
Free to use, modify, and redistribute. No attribution required.