isnadibot / plugins
AGENTS.md2 starsChanged 14 days agoMITView on GitHub
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/cursor-cloud.mdcCursor rule
- install-anti-slopSkill
- ios-debugger-agentSkill
- ios-simulator-browserSkill
- test-isnadi-appSkill
- test-isnadi-mobileSkill
The file
We split this AGENTS.md into the rules it gives an agent
8 rules. About 1 is common, 3 are distinctive, and 4 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 4
Use Custom MCP for experimental apps, wrappers, duplicates, and narrow utilities.
- Distinctive: line 8
Test add, connect, use, disable, re-enable, and remove behavior before acceptance.
- Distinctive: line 9
Keep PostgreSQL, SQLite, Redis, Docker, Playwright, time, fetch, filesystem, and generic memory servers in Custom MCP.
- Very specific: line 10
Keep one plugin in each entries/<id>/ directory.
- Very specific: line 15
Use verification-pending when a connection recipe exists but the real lifecycle has not passed.
- Very specific: line 16
Use approval-pending with an explicit blocker when a vendor must approve access.
- Common: line 17
Keep tokens and secrets in the host environment or service sign-in flow.
- Very specific: line 19
Run vp run plugins:check and the focused catalog tests after each entry change.
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md 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
Plugin catalog
- This is a curated directory, not a complete MCP registry. A valid manifest does not guarantee acceptance.
- Use Custom MCP for experimental apps, wrappers, duplicates, and narrow utilities.
- Admit a plugin only when people actively use the product and a user or maintainer requested the integration.
- Require a real IsnadiBot job, an official or trusted MCP server, and an accountable publisher.
- Require current setup and reference documentation, official logo provenance, and licensing.
- Test add, connect, use, disable, re-enable, and remove behavior before acceptance.
- Keep PostgreSQL, SQLite, Redis, Docker, Playwright, time, fetch, filesystem, and generic memory servers in Custom MCP.
- Keep one plugin in each
entries/<id>/directory. - Each directory contains
plugin.json,logo.svg, andlogo-dark.svg. Do not add generated catalog files. - Preserve an existing plugin
id. Its MCP server ID remainsbuiltin-<id>. - Record the current publisher, documentation, logo source, license, and connection health. A valid manifest does not guarantee directory acceptance.
- Mark a plugin
availableonly after its real add, connect, use, disable, re-enable, and remove lifecycle passes in IsnadiBot. - Use
verification-pendingwhen a connection recipe exists but the real lifecycle has not passed. - Use
approval-pendingwith an explicit blocker when a vendor must approve access. - Keep tokens and secrets in the host environment or service sign-in flow. Never put them in a manifest or ask an MCP server to export a provider token or secret.
- Declare an approval for every applicable send, pay, delete, production, secrets, publishing, signatures, refunds, and account-wide action.
- Run
vp run plugins:checkand the focused catalog tests after each entry change.
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.

