sol-validator-skills
AGENTS.md3 starsChanged 49 days agoMITView on GitHub
This repository owns reusable validator skills. Operator inventory, profiles, credentials, signer paths and operation records belong outside it. Configuration does not authorize transactions, restarts or identity changes. Author instructions and helpers in src/skills. Each skill's instructions.md becomes its published SKILL.md; supporting references are copied with rewritten local links. skills/<name> is the generated, independently installable Agent Skills distribution. Run bun run skills:build after source changes; do not hand-edit generated files. Shared runtime dependencies and their pinned lock are maintained in runtime/.
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 31,700 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 repository15 other files it gives its agents.
- doublezero-statusSkill
- inventorySkill
- jito-bam-boostSkill
- jitosol-stewardSkill
- marinade-bond-sweepSkill
- onboardingSkill
- sfdp-checkSkill
- upgrade-agave-primarySkill
- upgrade-agaveSkill
- upgrade-mainnetSkill
- upgrade-testnetSkill
- validator-performanceSkill
- validator-revenueSkill
- votex-roi-simSkill
- votex-statusSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 1 is common, 0 are distinctive, and 0 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈0
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
- could be automatic: a setting could enforce it
- Common: line 14
Use synthetic fixtures, never production mutations, as tests.
≈ “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
Public skill source
This repository owns reusable validator skills. Operator inventory, profiles, credentials, signer paths and operation records belong outside it. Configuration does not authorize transactions, restarts or identity changes.
Author instructions and helpers in src/skills. Each skill's instructions.md becomes its published SKILL.md; supporting references are copied with rewritten local links. skills/<name> is the generated, independently installable Agent Skills distribution. Run bun run skills:build after source changes; do not hand-edit generated files. Shared runtime dependencies and their pinned lock are maintained in runtime/.
Run bun run validate for offline regression tests, TypeScript, generated-file consistency, manifest/reference checks and the release allowlist. Run bun run test:install when packaging or installation changes; it uses the pinned skills CLI in a temporary project and does not install user/global skills. Network package installation precedes offline checks. Use Bun/TypeScript and the repository's .venv/bin/python for Python work.
Use synthetic fixtures, never production mutations, as tests. Do not read signer contents or commit RPC credentials. Respect operator-selected configuration paths and user-level defaults independently of the session directory. Keep approval requirements tied to actual mutations and preserve established live-role/tower checks.
Maintain Markdown created and last_updated dates, including string-valued skill metadata dates. Use Conventional Commits and split unrelated scopes. Keep release-files.json explicit: review new source files and generated outputs before adding them. Publishing commits, tags or release artifacts is a separate action from local preparation.
Apply DRY to authored sources: keep common operational policy and identical runtime behavior in src/skills/shared, and reuse helpers or linked references. Keep workflow-specific approval, network, missing-data and tower semantics explicit. Generated copies in skills/ are intentional for independent installation; consolidate their sources and regenerate them.
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.