seri-agent
AGENTS.md0 starsChanged 2 months agoApache-2.0View on GitHub
Guidance for AI agents working in this repository. This repository is the seri CLI harness. It ships as the seri binary, written in TypeScript on Bun. Work here is apps/cli and the packages that binary needs (packages/daemon-client, packages/model-catalog, packages/plans). Hosted apps, billing, and research docs live in lzvxck/seri-agent. Do not add them here. - bun run dev -- <args> — run the CLI from source - bun test — run the test suite (bun's built-in runner) - bun test path/to/file.test.ts…
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 33,315 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 repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- .cursor/rules/code-quality.mdcCursor rule
- .cursor/rules/git-workflow.mdcCursor rule
- .cursor/rules/hooks.mdcCursor rule
- .cursor/rules/pstack.mdcCursor rule
- perf-reviewSkill
- verify-seriSkill
- perf-reviewSkill
- thermo-nuclear-code-quality-reviewSkill
The file
We split this AGENTS.md into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Very specific: line 26
Write tools are gated via WRITE_TOOL_NAMES in apps/cli/src/provider/tools.ts.
- Very specific: line 30
Do not hand-edit apps/cli/src/tools/rg-vendored.bin.
≈ “Common” is an estimate: at least 1 in 5 of the 42,283 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
AGENTS.md
Guidance for AI agents working in this repository.
What this is
This repository is the seri CLI harness. It ships as the seri binary, written in
TypeScript on Bun. Work here is apps/cli and the packages that binary needs
(packages/daemon-client, packages/model-catalog, packages/plans).
Hosted apps, billing, and research docs live in lzvxck/seri-agent. Do not add them here.
Commands
bun run dev -- <args>— run the CLI from sourcebun test— run the test suite (bun's built-in runner)bun test path/to/file.test.ts— run a single test filebun run typecheck(aliaslint) —tsc --noEmitbun run build— compile todist/serifor the current platform- CI (
.github/workflows/ci.yml) runs typecheck + test + build on Linux, macOS, and Windows on every push — treat all three as required, not just the local OS
Invariants
- The loop is a library:
apps/cli/src/loop/loop.ts. It never writes stdout. - Write tools are gated via
WRITE_TOOL_NAMESinapps/cli/src/provider/tools.ts. bash/powershellnever get an "always" grant.memory_writeis the deliberate exception; it is not in that list.- Conductor is pstack (
/poteto-mode). Do not run a second orchestrator. Do not merge. - Do not hand-edit
apps/cli/src/tools/rg-vendored.bin.
Open the file that owns a behavior instead of restating it here. Cancellation lives in
apps/cli/src/signals.ts. The gate lives in apps/cli/src/gate/gate.ts. After
--help/--version/--selftest, argv launches seri serve, seri exec, seri doctor,
seri update, and a task (or a TTY idle TUI). -- is the escape that makes those words a task.
Review
Prioritize the loop, tools, permissions, and cancellation. Prefer the smallest change that matches existing tests. Feature work lands via a branch and a PR.
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.

