wraith-skills
AGENTS.md2 starsChanged 6 days agoMITView on GitHub
wraith is a plugin of skills and agents for data and infrastructure engineering on GCP. It serves Claude Code and GitHub Copilot from one repository. - uvx --python 3.12 pre-commit run --all-files (lint, copilot check, privacy scan, file types and plugin checks, all from wraith-tooling) - uvx --from git+https://github.com/wraith-proxy/wraith-tooling@ee8df52790af86f605d0653bd588ccd1be2cbb6b wraith-build-copilot - claude plugin validate --strict . - claude plugin validate --strict plugin - claude plugin eval . - Portable skill frontmatter only: name, description, license, compatibility, metadata, allowed-tools. - The…
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 31,439 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 repositoryOne other file it gives its agents.
- .claude/CLAUDE.mdCLAUDE.md
The file
We split this AGENTS.md into the rules it gives an agent
5 rules. About 1 is common, 1 is distinctive, and 3 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
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
- Distinctive: line 23
No em dash (U+2014) in skills, agents, hooks, evals, README or AGENTS.md.
- Very specific: line 24
Never edit plugin/com.github.copilot/agents/.
- Very specific: line 26
Commit format 👻 <type>/WSK-<5 digits>: <summary>, types feature, fix, bug, break (these four release) and docs, chore, ci, test, refactor …
- Common: line 31
No private data: personal emails, private IPs, internal hostnames, home paths, employer or client names, credentials.
- Very specific: line 33
Only plugin content lives in plugin/ (manifests, skills, agents, the icon, the listing README and legal files).
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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
wraith-skills
wraith is a plugin of skills and agents for data and infrastructure engineering on GCP. It serves Claude Code and GitHub Copilot from one repository.
Commands
uvx --python 3.12 pre-commit run --all-files(lint, copilot check, privacy scan, file types and plugin checks, all from wraith-tooling)uvx --from git+https://github.com/wraith-proxy/wraith-tooling@ee8df52790af86f605d0653bd588ccd1be2cbb6b wraith-build-copilotclaude plugin validate --strict .claude plugin validate --strict pluginclaude plugin eval .
Rules
- Portable skill frontmatter only:
name,description,license,compatibility,metadata,allowed-tools. - The plugin ships only Markdown, text, JSON, YAML and images
(
wraith-file-typesenforces it). Tooling lives in wraith-proxy/wraith-tooling. - No em dash (U+2014) in skills, agents, hooks, evals, README or AGENTS.md.
- Never edit
plugin/com.github.copilot/agents/; it is generated bywraith-build-copilot. - Commit format
👻 <type>/WSK-<5 digits>: <summary>, typesfeature,fix,bug,break(these four release) anddocs,chore,ci,test,refactor(no release). No attribution trailers, never--no-verify. PR titles use the same format: the squash subject decides the release, and CI rejects a title that does not match. - No private data: personal emails, private IPs, internal hostnames, home paths, employer or client names, credentials.
- Only plugin content lives in
plugin/(manifests, skills, agents, the icon, the listing README and legal files);wraith-check-pluginenforces it. Evals, docs and CI stay at the repo root.
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.

