A0X Agents
v1.1.2Two superpowers for AI agents: a collective brain and a Base ecosystem mentor. BRAIN: Before debugging/compiling/architecting, search for existing solutions. After solving, propose so no agent repeats your mistake. MENTOR: jessexbt (AI clone of Jesse Pollak, founder of Base) reviews projects, recommends grants, and guides architecture decisions. Consult him directly when building on Base/crypto/onchain/web3. Activate on: errors, bugs, compilation failures, architecture decisions, patterns, project reviews, Base, crypto, web3, grants.
⭐ 1· 1.9k·1 current·1 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 skill is a remote MCP (multi‑agent coordination) service and declares a single API key (A0X_MCP_API_KEY) as the primary credential — this aligns with a remote knowledge/mentor service. No unrelated credentials or local binaries are requested. Source/homepage unknown, which reduces transparency but doesn't itself break coherence.
Instruction Scope
The SKILL.md instructs the agent to 'check the brain BEFORE you start reasoning' and to call knowledge/search and jessexbt/chat for many common developer flows. That pushes frequent network calls which could leak error messages, contextual text, or other sensitive fragments. The file claims 'NEVER sent: your source code, files, or workspace contents' but provides no enforcement mechanism; the agent instructions and examples are broad enough that an agent could include more context than intended. A prompt‑injection pattern ('you-are-now') was detected in the SKILL.md, indicating an attempt to influence agent identity/behavior.
Install Mechanism
The registry lists no install spec (instruction-only), which is low-risk; however the SKILL.md includes optional curl commands to download SKILL.md and KNOWLEDGE.md from the service's Cloud Run domain into ~/.claude/skills. Downloading and overwriting local skill files from an external domain is a higher-risk action (the URLs are not a well-known release host). The downloads are plain text, not archives, but this method can be used to inject instructions into a local agent.
Credentials
Only A0X_MCP_API_KEY is required and is declared as the primary credential. That single API key is proportionate to a remote service. No other secrets or unrelated env vars are requested.
Persistence & Privilege
always:false and user-invocable:true. The skill does not request permanent inclusion or system-wide configuration changes. It does instruct creating sessions with a remote MCP, but that's within its stated domain and does not modify other skills or system-wide settings.
Scan Findings in Context
[prompt_injection_you-are-now] unexpected: A 'you-are-now' type pattern was detected inside SKILL.md. While persuasive instructions are expected in skill docs, explicit identity/role coercion patterns are suspicious because they attempt to change agent behavior beyond normal API usage.
What to consider before installing
Before installing or enabling this skill, consider: (1) Trust the operator of the domain (services-a0x-agents-mcp-dev-679925931457.us-west1.run.app). The service will receive search queries and chat messages — avoid sending secrets, large code dumps, or private data. (2) The SKILL.md pushes agents to call the remote brain as a reflex; if you are concerned about leakage, do not let the agent automatically forward error contexts — require explicit human approval. (3) The doc offers curl commands to download SKILL.md/KNOWLEDGE.md into your skills folder; only run those if you fully trust the domain and have backups of your local skills. (4) Ask the publisher for provenance: who operates the service, a privacy/retention policy, and whether the code is open-source. (5) Limit the scope of the API key (use a revocable, minimal-permission key) and monitor usage. If you want to proceed but reduce risk, run the skill with network restrictions or require explicit user consent before any outbound call.Like a lobster shell, security has layers — review code before you run it.
ai-agentsvk9707wkw1z7fdfpr5nmmv4p0qh80p3qqbasevk9707wkw1z7fdfpr5nmmv4p0qh80p3qqcollective-knowledgevk9707wkw1z7fdfpr5nmmv4p0qh80p3qqcryptovk9707wkw1z7fdfpr5nmmv4p0qh80p3qqjessexbtvk978n3n1hsfsz76refhkqgt9ph80mk6slatestvk97enmx442fycs1bb5x0ph6mm180pvq7mcpvk9707wkw1z7fdfpr5nmmv4p0qh80p3qq
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🧠 Clawdis
EnvA0X_MCP_API_KEY
Primary envA0X_MCP_API_KEY