agent-tool-design
Skill1 starsChanged 2 months agoMITView on GitHub
Design or refactor tool contracts shared by an agent SDK, MCP server or CLI. Use when schemas drift between transports, completion is ambiguous, or tool discovery and output size need redesign. Not a general agent framework selection guide.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- mattstruble/skillstheir own version, first seen Jul 22, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- mattstruble/skills★ 2same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository2 other files it gives its agents.
- AGENTS.mdAGENTS.md
- strands-agentsSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Read contract and completion design for registry boundaries, terminal actions, discovery, images and transport parity.
- Distinctive: line 22
Keep the registry independent of a particular agent SDK where practical.
- Distinctive: line 25
Test what a real client discovers and invokes.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
One tool contract, deliberate transport differences
Identify the caller, state owner and completion owner before changing a tool surface. List the actual transports in use. Preserve existing public contracts unless the user authorizes their evolution.
Read contract and completion design for registry boundaries, terminal actions, discovery, images and transport parity.
Workflow
- Choose one source of truth for name, description, input schema and execution. Keep the registry independent of a particular agent SDK where practical.
- Let transport adapters translate that definition. Document intentional differences: authentication, result encoding, cancellation and completion ownership.
- Test what a real client discovers and invokes. A test comparing two projections of the same helper can pass while the public transport is broken.
- Make completion and artifact identity observable. Identify the saved revision or result, not merely a successful tool call or the model saying “done.”
- Return errors that suggest a valid next action. Distinguish unsupported capability, temporary failure, invalid input and an uncertain side effect.
For a large surface, prefer bounded search and describe results over placing every schema in the initial prompt. Keep authorization in execution even if discovery filters capabilities. A compact discovery surface is not an authorization boundary.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.