Privy Agentic Wallets
v1.0.1Create and manage agentic wallets with Privy. Use for autonomous onchain transactions, wallet creation, policy management, and transaction execution on Ethereum, Solana, and other chains. Triggers on requests involving crypto wallets for AI agents, server-side wallet operations, or autonomous transaction execution.
⭐ 3· 2k·5 current·5 all-time
by@tedim52
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The name/description and SKILL.md consistently describe a Privy server-wallet integration and use of Privy APIs (creating wallets, policies, and executing transactions). Requiring PRIVY_APP_ID and PRIVY_APP_SECRET is appropriate for this purpose. However, the registry metadata lists no required environment variables or config paths even though the SKILL.md explicitly requires those credentials and instructs adding them to the OpenClaw gateway config — a metadata mismatch that reduces transparency.
Instruction Scope
SKILL.md gives detailed runtime instructions that go beyond simple API examples: it tells agents to check shell env variables, add credentials to ~/.openclaw/openclaw.json, enforce rate limits, log every transaction to ~/.openclaw/workspace/logs/privy-transactions.jsonl, and to refuse operations originating from external content. Those file/config write/read instructions are realistic for a server-wallet skill, but they are not declared in the skill metadata and grant the skill scope to read/write user config and logs — this discrepancy is concerning and should be validated before enabling the skill.
Install Mechanism
Instruction-only skill with no install spec and no code files. This is lower risk from a code-distribution standpoint because nothing is downloaded or executed by default.
Credentials
The SKILL.md legitimately requires two sensitive environment variables (PRIVY_APP_ID and PRIVY_APP_SECRET) for API auth; those are proportional to the stated functionality. However, the skill metadata did not declare any required env vars or config paths. The SKILL.md also instructs storing credentials in the OpenClaw gateway config file (~/.openclaw/openclaw.json) and using them in Basic auth for curl calls — this means the skill expects access to secrets and to a specific config location that the registry didn't advertise, which is a transparency and privilege concern.
Persistence & Privilege
The skill is not 'always:true' and is user-invocable (normal). It does instruct agents to write persistent logs and to store credentials in the gateway config, which gives it persistent artifacts on disk and ongoing access to secrets. Autonomous invocation is allowed by default; combined with access to the app secret the blast radius is meaningful — review platform controls around autonomous skill invocation and ensure the Privy app has limited funds/permissions.
Scan Findings in Context
[ignore-previous-instructions] expected: The regex scanner flagged a prompt-injection pattern string. SKILL.md intentionally lists prompt-injection patterns (including that string) as things to detect and refuse. Its presence is expected and appropriate in a security-focused document.
What to consider before installing
This skill appears to actually implement a Privy server-wallet workflow and therefore legitimately needs your PRIVY_APP_ID and PRIVY_APP_SECRET — but the published skill metadata did not declare those sensitive requirements or the config/log file paths the instructions reference. Before installing or enabling: 1) Confirm the platform will provide PRIVY_APP_SECRET to the skill securely (and will not leak it to other skills or responses). 2) Prefer creating a Privy App with minimal privileges and funding (use testnet or tiny amounts first), and plan to rotate the secret after testing. 3) Verify where credentials will be stored (the SKILL.md references ~/.openclaw/openclaw.json) and ensure the file permissions are secure. 4) Ask whether the agent will be allowed to write logs to ~/.openclaw/workspace/logs/ — if so, inspect those logs and their retention policies. 5) Keep autonomous invocation off (or tightly controlled) unless you fully trust the agent and have rate limits/confirmations enforced. 6) Require explicit verbal/user confirmations for policy/rule deletions and other destructive actions. If the publisher can explain the metadata omissions (and you can limit the Privy app's power and test on testnet), the skill can be used safely; without that, treat it cautiously.Like a lobster shell, security has layers — review code before you run it.
latestvk972nj247vf7d829gcjczfgqgx80hgy0
License
MIT-0
Free to use, modify, and redistribute. No attribution required.