requie/endostack
v1.0.1MIT
Security-first agent workflows for coding agents: route work through tested playbooks, review with multiple subagents, and verify before you ship.
scripts
- show-me-your-work
- scripts/log.mjs
- scripts/test/log.test.mjs
- statechart
- scripts/lib/chart.mjs
- scripts/lib/config.mjs
- scripts/lib/dot.mjs
- scripts/lib/files.mjs
- scripts/lib/graph.mjs
- scripts/lib/mermaid.mjs
73 more
- scripts/lib/pipeline.mjs
- scripts/lib/release.mjs
- scripts/lib/replay.mjs
- scripts/lib/report.mjs
- scripts/lib/sany.mjs
- scripts/lib/tlc-output.mjs
- scripts/lib/toolchain.mjs
- scripts/lib/values.mjs
- scripts/lib/verdicts.mjs
- scripts/lib/view.mjs
- scripts/statechart.mjs
- scripts/test/chart.test.mjs
- scripts/test/cli.test.mjs
- scripts/test/config.test.mjs
- scripts/test/corpus.test.mjs
- scripts/test/corpus/commit/view.mjs
- scripts/test/corpus/login-regions/view.mjs
- scripts/test/corpus/login/view.mjs
- scripts/test/corpus/merge/view.mjs
- scripts/test/corpus/player/view.mjs
- scripts/test/graph.test.mjs
- scripts/test/index.js
- scripts/test/live.test.mjs
- scripts/test/support/corpus.mjs
- scripts/test/support/record.mjs
- scripts/test/tlc-output.test.mjs
- scripts/test/values.test.mjs
- scripts/test/view.test.mjs
- tarique-mode
- scripts/check-plan.mjs
- scripts/doctor.mjs
- scripts/orch/lib/cli.mjs
- scripts/orch/lib/errors.mjs
- scripts/orch/lib/files.mjs
- scripts/orch/lib/frontier.mjs
- scripts/orch/lib/gates.mjs
- scripts/orch/lib/lock.mjs
- scripts/orch/lib/store.mjs
- scripts/orch/orch.mjs
- scripts/orch/test/cli.test.mjs
- scripts/orch/test/frontier.test.mjs
- scripts/orch/test/helpers.mjs
- scripts/orch/test/store.test.mjs
- scripts/run-tests.mjs
- scripts/test/check-plan.test.mjs
- scripts/test/doctor.test.mjs
- scripts/test/worktree-audit.test.mjs
- scripts/watch-pr/lib/args.mjs
- scripts/watch-pr/lib/classify.mjs
- scripts/watch-pr/lib/config.mjs
- scripts/watch-pr/lib/context.mjs
- scripts/watch-pr/lib/errors.mjs
- scripts/watch-pr/lib/exec.mjs
- scripts/watch-pr/lib/github.mjs
- scripts/watch-pr/lib/main.mjs
- scripts/watch-pr/lib/proof.mjs
- scripts/watch-pr/lib/render.mjs
- scripts/watch-pr/lib/snapshot.mjs
- scripts/watch-pr/lib/verdicts.mjs
- scripts/watch-pr/lib/watch.mjs
- scripts/watch-pr/test/args.test.mjs
- scripts/watch-pr/test/context.test.mjs
- scripts/watch-pr/test/fakes.mjs
- scripts/watch-pr/test/proof.test.mjs
- scripts/watch-pr/test/snapshot.test.mjs
- scripts/watch-pr/test/watch.test.mjs
- scripts/watch-pr/watch-pr.mjs
- scripts/worktree-audit.mjs
- try-endostack
- demo/app/server.mjs
- demo/app/test/server.test.mjs
- demo/check.mjs
- scripts/test/demo.test.mjs
plugin.json
- name
- endostack
- version
- 1.0.1
- description
- Security-first agent workflows for coding agents: route work through tested playbooks, review with multiple subagents, and verify before you ship.
- author
- Tarique Smithhttps://github.com/requie/endostack
- license
- MIT
- repository
- https://github.com/requie/endostack
- extensions
- com.openai
plugin.json
{
"$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
"name": "endostack",
"version": "1.0.1",
"description": "Security-first agent workflows for coding agents: route work through tested playbooks, review with multiple subagents, and verify before you ship.",
"author": {
"name": "Tarique Smith",
"url": "https://github.com/requie/endostack"
},
"homepage": "https://github.com/requie/endostack",
"repository": "https://github.com/requie/endostack",
"license": "MIT",
"keywords": [
"agent-skills",
"security",
"code-review",
"subagents",
"workflow",
"verification"
],
"extensions": {
"com.openai": {
"onboardingSkill": "./skills/setup-endostack/SKILL.md",
"interface": {
"displayName": "EndoStack",
"shortDescription": "Security-first agent workflows",
"longDescription": "EndoStack makes the agent read code before changing it, compare more than one design, and check its result by running what a user would run. When a change touches a trust boundary, such as input parsing, file paths, auth or secrets, it threat-models the change before building it and runs a security review on the diff before shipping. Text the agent reads from issues, web pages and logs is treated as data, never as instructions.\n\nType $try-endostack for a five-minute demo: the agent fixes a bug whose obvious one-line fix would leak the demo app's .env file, and a check script shows the difference. Type $tarique-mode with an issue link or a one-sentence bug description to use the same workflow on your own code.\n\nEndoStack collects no data and has no telemetry. Its scripts run on your machine.",
"developerName": "Tarique Smith",
"category": "Developer Tools",
"capabilities": [
"Read",
"Write"
],
"websiteURL": "https://github.com/requie/endostack",
"supportURL": "https://github.com/requie/endostack/issues",
"privacyPolicyURL": "https://github.com/requie/endostack/blob/main/docs/privacy.md",
"termsOfServiceURL": "https://github.com/requie/endostack/blob/main/LICENSE",
"defaultPrompt": [
"Run $try-endostack for the five-minute demo",
"Threat model the changes on this branch before I merge",
"Run a security review on this branch's diff"
],
"brandColor": "#0F172A",
"composerIcon": "./assets/logo.png",
"logo": "./assets/logo.png"
}
}
}
}
What else this package ships
These files come with the package and this site does not publish them. They are listed so you know what is there before you install it.
- 79JavaScript files
- 101JSON files
- 30text files
- 26configuration files
- decisions.template.tsv
- 84other files
Client extensions
Data this package carries for particular clients. The directory lists the clients named and never reads what is addressed to them.
- com.github.copilotships a directory of files
- com.openai