NetFabric/NetFabric.Numerics / .agents/skills/agents-md
agents-md
VerifiedAuthor and maintain AGENTS.md files — the open, tool-agnostic format for giving coding agents project context. Use when: creating AGENTS.md; writing an agent instructions file; root or nested AGENTS.md for monorepos; migrating CLAUDE.md, AGENT.md, .cursor/rules, or copilot-instructions.md to AGENTS.md; greenfield project setup before code exists; brownfield onboarding for an existing codebase; documenting build/test/lint commands, code style, PR/commit conventions, and security considerations for agents; AGENTS.md discovery and precedence rules. Works for any language, framework, or project size. DO NOT USE FOR: authoring SKILL.md skill files (use create-skill); APM package/plugin authoring (use apm-producer); language-specific style rules (use language skills e.g. csharp-best-practices).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.4 KB |
| references/anatomy-and-discovery.md | markdown | 3.2 KB |
| references/authoring-workflow.md | markdown | 4.5 KB |
| references/quality-and-security.md | markdown | 2.2 KB |
| references/writing-style.md | markdown | 2.0 KB |
references/quality-and-security.md
View on GitHub# Quality & Security
See [writing-style.md](writing-style.md) for wording and formatting conventions.
## Verification Is Mandatory
Agents will execute testing/build/lint commands found in `AGENTS.md` automatically and attempt to fix failures before finishing a task. Every listed command must:
- Actually succeed on a clean checkout — never transcribe a command from stale docs without running it.
- Be safe to run unattended and non-interactively (no prompts requiring manual input).
- Be idempotent or clearly scoped (a "reset DB" command listed casually will get run by an agent).
## Security Considerations to Cover
Include a Security section whenever any of these apply (most repos):
- **No secrets in the file** — `AGENTS.md` is typically committed and world-readable; never place API keys, tokens, or credentials in it, even as examples.
- **Flag destructive commands** — migrations, force-push, prod deploys, data-deleting scripts — require explicit confirmation before an agent runs them; state that explicitly rather than assuming caution.
- **Note sandbox/network limits** — if agents in this environment run without network access or in a restricted sandbox, say so, so agents don't waste turns retrying blocked calls.
- **Protected paths** — call out generated code, vendored dependencies, or lockfiles that must never be hand-edited by an agent.
## Maintenance
- Update `AGENTS.md` in the **same PR** that changes build, test, or lint tooling — treat a stale instruction as a bug, not a documentation nit.
- Periodically prune commands/sections referencing tools or scripts that no longer exist.
- Re-validate nested files after restructuring a monorepo (package moved, renamed, or merged).
## Anti-Patterns
- ❌ Committing secrets, tokens, or connection strings "for convenience"
- ❌ Listing commands that don't currently pass
- ❌ Baking language- or framework-specific advice into a supposedly project-agnostic template
- ❌ Leaving placeholder/`TODO` text in a committed `AGENTS.md`
- ❌ Symlinking `CLAUDE.md` to `AGENTS.md` instead of using the `@AGENTS.md` import directive
- ❌ Copying `AGENTS.md` content into `CLAUDE.md` instead of importing it, creating two documents to keep in sync
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |