- Home
- Skills
- Agents & Orchestration
- toolsmith
Works with the AI Agents you already use
toolsmith
Find, audit, and install secure, version-locked agent skills tailored to your project stack.
$5.99
toolsmith
Example session with this skill installed
Check my current stack and see if there are any missing skills for the React frontend and Node backend layers.
- Read your context and instructions
- Compiled the toolsmith
I have analyzed the codebase and identified gaps in your current agent capabilities.
Stack Audit and Needs Assessment
| Layer | Identified Need | Status |
|---|---|---|
| Frontend | React/Tailwind component consistency | Missing |
| Frontend | Zod schema validation patterns | Missing |
| Backend | Prisma migration and safety rules | Installed (v1.2.0) |
| Backend | Express middleware error handling | Missing |
I have found 3 potential candidates for the React/Tailwind need from your allowed sources:
ui-standard-skill(High signal, 420 stars)tailwind-refactor-pro(Medium signal, pinned version available)react-atomic-design(Low signal)
Next steps
- Select a candidate number to begin the quarantined fetch.
- Add
github.com/trusted-dev/skillsto your allowlist to see more results. - Use the
Generatecommand to draft a custom Zod skill based on yoursrc/schemasfolder.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing agent capabilities at scale often leads to "prompt rot" or the accidental execution of untrusted third-party instructions. Toolsmith solves the discovery and security problem by providing a formal pipeline for finding, vetting, and installing agent skills into your project.
What it does
- Stack Analysis reads your manifests and source code to identify exactly which skills your agents are missing.
- Secure Discovery searches trusted sources for existing skills that match your specific tech stack.
- Quarantined Auditing clones candidates into isolated temporary directories for static analysis before you ever see the code.
- Injection Defense runs an automated audit to detect prompt injection or malicious instructions hidden in third-party skills.
- Skill Generation authors custom, project-specific skills from scratch when no trustworthy third-party version exists.
How it works
- Identify the project stack and current skill gaps across your application layers.
- Search and select candidates from allowed repositories or local folders.
- Vet the candidate via a two-stage audit involving static checks and isolated reading.
- Install the approved skill to a locked directory with a pinned commit hash.
Frameworks & tools
This skill requires a shell environment with git, grep, find, and tar. It works across macOS, Linux, and Git Bash on Windows, integrating with any agentic workflow that utilizes a SKILL.md or similar instruction format.
Why this beats prompting it yourself
Manual prompt engineering lacks version control and security vetting. This skill ensures every instruction set in your codebase is pinned to a specific commit, audited for safety, and explicitly mapped to your actual file structure.
Use cases
- Auditing a third-party agent skill before adding it to a production repo.
- Generating custom instructions for a legacy codebase with unique patterns.
- Keeping a
skills-lock.mdfile to sync agent capabilities across a dev team.
Known limitations
Requires local shell access for git and file operations. It cannot execute web-based installers or fetch via branch names to ensure security pins remain valid.
Requires a shell with git, grep, find, and tar. Cannot use web-fetch tools or branch-based installs. Does not support non-Git version control.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 days ago
Requires