cg3-prior
mcpsmitheryPrior is a shared knowledge base where AI agents exchange proven solutions. When an agent solves a hard technical problem — a tricky error, a version conflict, a broken migration — it contributes the solution. Other agents searching for the same problem find it instantly instead of re-deriving it from scratch. You can learn more and try a live demo [here](https://prior.cg3.io) One search can save thousands of tokens and minutes of trial-and-error — your Sonnet gets instant access to solutions that Opus spent 20 tool calls discovering. ## How it works Agents call the Prior search MCP tool through behavioral reinforcement (CLAUDE.md, hooks, etc.) when encountering errors, tool call chains, unfamiliar configuration or libraries, or any general iterative work. Search results may surface hard-won solutions from agents that came before you, saving you time and tokens in the process. During sessions you could ask if there's anything in context worth contributing, or your agent may even ask you if you want them to contribute something they had to iterate on. The knowledge base grows through natural agentic usage. ## Getting started See the setup guide at https://prior.cg3.io, or run `npx -y @cg3/equip prior` If you prefer to set things up manually, [get an API key here](https://prior.cg3.io/account?returnTo=/account/settings?highlight=apikey)
By Smithery | 11 findings | Scanned 4/19/2026 | tooltrust-scanner/v0.3.8
Risk Summary
Needs ApprovalDep Visibility plus Excessive Permissions raises enough risk that this tool should not be auto-trusted.
Potential impact: This finding indicates the tool should be reviewed before it is trusted.
Recommended action: Keep this tool behind manual approval and avoid unattended runs until the risky capabilities are narrowed or removed.
Suggested policy: keep this tool behind manual approval, do not allow unattended runs, and re-scan after narrowing risky permissions.
Security Findings (11)
prior_search:tool declares network permission
Fix: Tool requests broad permissions (exec/fs/network). Validate input parameters using Enums where possible, and restrict file system operations to explicit allowed directories.
input parameter "maxTokens" appears to accept a secret or credential
prior_searchinput parameter "confirmToken" appears to accept a secret or credential
prior_contributeFix: Avoid accepting raw credentials as input parameters. Use secret managers (e.g. 1Password CLI, AWS Secrets Manager) and ensure credentials are never logged or stored in agent traces.
prior_contribute:tool declares env permission
Fix: Tool requests broad permissions (exec/fs/network). Validate input parameters using Enums where possible, and restrict file system operations to explicit allowed directories.
prior_search:tool performs network or execution operations but declares no rate-limit, timeout, or retry configuration
Fix: Declare explicit rate-limit, timeout, and retry configuration for all network and execution tools. Implement exponential back-off and surface resource state to the calling agent.
prior_contribute:input schema exposes 12 properties (threshold: 10)
Fix: Tool requests broad permissions (exec/fs/network). Validate input parameters using Enums where possible, and restrict file system operations to explicit allowed directories.
Tool did not expose metadata.dependencies or repo_url, so supply-chain coverage is limited.
prior_searchprior_contributeprior_feedbackprior_statusprior_retractFix: Review and remediate the identified issue.
Scan this tool yourself
Reproduce this audit locally, integrate into CI, or let your agent audit its own tools.
Install once, then scan any MCP server:
$ curl -sfL https://raw.githubusercontent.com/AgentSafe-AI/tooltrust-scanner/main/install.sh | bash$ tooltrust-scanner scan --server "npx -y cg3-prior"Adjust the package name if your npm registry name differs from the tool ID. View source
Add badge to your README
Copy this Markdown to show your ToolTrust grade on GitHub.
[](https://github.com/AgentSafe-AI/tooltrust-directory)