β³ This skill is pending AI review.
Scores will appear once the review pipeline completes.
compile-basic
Basic compiler conformance fixture.
Choose how to use this skill
You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.
1. Native installer
This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.
Do not guess an installer command or replace an existing version without reviewing the diff.
2. Complete package recommended
Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.
No complete ProSkills package is published for this listing yet.3. Prompt-only
Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.
Need only the instruction file?
Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.
No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.
// RATINGS
Not yet listed on ClawHub or SkillsMP
// README
π MDA Open Spec β Markdown for Agent
A Markdown superset for agent-facing documents. One source, many targets β compile into the
.mdfiles every major agent runtime already loads. Tamper-evident at load β every artifact carries a reproducible content digest, and signed artifacts carry Sigstore-anchored signatures, so neither the agent loading the doc nor the human reviewing it has to trust an unsigned blob.
Read in other languages: English Β· δΈζ Β· Deutsch Β· EspaΓ±ol Β· FranΓ§ais Β· Π ΡΡΡΠΊΠΈΠΉ Β· νκ΅μ΄ Β· ζ₯ζ¬θͺ Β· ΰ€Ήΰ€Ώΰ€¨ΰ₯ΰ€¦ΰ₯
What MDA is
Until now, you shipped the same skill four times. Once as SKILL.md for the agentskills.io runtimes. Once as AGENTS.md for the AAIF ecosystem. Once as MCP-SERVER.md with a sidecar JSON. Once as CLAUDE.md. Same content, four frontmatter shapes. Update one, forget the others, and a month in, the four files have quietly drifted into four slightly different instruction files.
You write one .mda. The compiler emits the rest.

βββββββββββββββββββββββββββ
β <name>.mda (source) β β MDA superset
ββββββββββββββ¬βββββββββββββ
β mda compile
βΌ
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β <name>/SKILL.md (+ scripts/, references/, assets/) β
β AGENTS.md β
β <name>/MCP-SERVER.md (+ mcp-server.json sidecar) β
β CLAUDE.md β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
drop-in compatible
Compatibility
Write your skill once. It drops into every major agent runtime β unchanged.
Five SKILL.md runtimes verified end-to-end with reproducible install kits in compat/. Compiled AGENTS.md artifacts also drop into the AAIF ecosystem (Codex, Copilot, Cursor, Windsurf, Amp, Devin, Gemini CLI, VS Code, Jules, Factory), and at the documentation level into skills.sh and other 2026 SKILL.md consumers.
And those four files can't say who signed them. The agent loading SKILL.md has no way to verify the content matches what you wrote, and the curator reviewing AGENTS.md has no way to know whose hands have been on it between merge and load. The standard frontmatter shapes have nowhere to put a content digest or a signature, so the trust decision quietly falls back to "we trust the repo, somehow."
MDA carries a JCS-canonicalized integrity.digest and DSSE-enveloped, Sigstore-anchored signatures[] in the frontmatter itself. Both sides β the agent at load time and the human at review time β can make a real trust decision against the artifact in hand, not against a feeling about the repo. Tamper-evidence and signer verification ship in the contract, not as a later bolt-on.

.mda adds three things on top of standard Markdown. All of them optional.
- Rich YAML frontmatter. Beyond the open-standard
nameanddescriptionbaseline, MDA carriesdoc-id,version,requires,depends-on,relationships, andtags. Agent-aware tools use these for routing, dependency resolution, and graph traversal. Seespec/v1.0/02-frontmatter.mdandspec/v1.0/10-capabilities.md. - Typed footnote relationships. Standard Markdown footnotes whose payload is a JSON object:
parent,child,related,cites,supports,contradicts,extends. Mirrored tometadata.mda.relationshipsin body order on compile. Seespec/v1.0/03-relationships.md. - Cryptographic identity. A JCS-canonicalized
integritydigest plus DSSE-enveloped, Sigstore-anchoredsignatures[]. The compiled.mdcarries reproducible tamper detection without bolting it on later. Seespec/v1.0/08-integrity.mdandspec/v1.0/09-signatures.md.
A .mda source with only the open-standard frontmatter compiles unchanged into a .md. Use as much or as little of MDA as your project needs.
Why this exists
The honest version. I kept shipping the same skill four times. Same content, four wrappers. Each runtime had its own opinions about what frontmatter belonged at the top and what counted as vendor-specific. The third or fourth time I copy-pasted a paragraph between SKILL.md and AGENTS.md and then watched them drift, I started writing this.
The thing is, the duplication isn't the worst part. The worst part is what you can't say in any of those formats. You can't say "this skill depends on that one, version ^1.2.0, with this content digest." You can't say "this file was signed by this identity at this Rekor index." You can't say "the relationship between this document and that one is supports, not cites." There's nowhere to put that information, so it sits in prose, where neither agents nor humans can act on it reliably.
MDA puts those things in the frontmatter and footnotes, in shapes a JSON Schema can validate. The Markdown body still renders. The standard fields still load. Everything new is optional. That's the whole pitch.
For the long version, two documents go deeper. Both trace every claim back to a section of the spec, and both call out current ecosystem gaps inline. Read them if you're deciding whether to adopt.
docs/v1.0/ai-agent-core-value.mdβ five points framed for runtimes, harnesses, validators, and dispatchers. What MDA gives an agent at load time: structuredrequiresfor typed dispatch, verifiable trust at load, machine-readable graph edges, filename-based one-lookup target dispatch, and the same validation contract for agent-authored and compiler-emitted output.docs/v1.0/human-curator-user-core-value.mdβ six points framed for the people who write and curate agent-facing instruction libraries. What MDA gives an author at ship time: one source into multiple ecosystems, tamper-evidence and publisher attribution, machine-readable dependency graph and version pinning, LLM-mediated authoring without learning every runtime's frontmatter, smaller (not zero) vendor lock-in, and strict validation that catches almost-conformant artifacts before they ship.
Three authoring modes
MDA artifacts may be produced
// HOW IT'S BUILT
// FILES
5 files shipped with this skill. Inline text is the catalog copy. The upstream link is the source repository.
conformance/compile/46-basic-targets/expected/SKILL.md351 Bffde8fd4664bupstream β
--- name: compile-basic description: Basic compiler conformance fixture. allowed-tools: Read Bash(echo:*) metadata: snoai-llmix: module: search_summary preset: openai_fast common: provider: openai model: gpt-5-mini mda: requires: network: none --- # Compile basic Use this fixture to confirm target projection.conformance/compile/46-basic-targets/expected/AGENTS.md370 B46b0002e9eecupstream β
--- name: compile-basic description: Basic compiler conformance fixture. metadata: snoai-llmix: module: search_summary preset: openai_fast common: provider: openai model: gpt-5-mini claude-code: allowed-tools: Read Bash(echo:*) mda: requires: network: none --- # Compile basic Use this fixture to confirm target projection.conformance/compile/46-basic-targets/expected/mcp-server.json73 B98b372b3639fupstream β
{ "name": "compile-basic", "version": "0.1.0", "transport": "stdio" }conformance/compile/46-basic-targets/expected/MCP-SERVER.md370 B46b0002e9eecupstream β
--- name: compile-basic description: Basic compiler conformance fixture. metadata: snoai-llmix: module: search_summary preset: openai_fast common: provider: openai model: gpt-5-mini claude-code: allowed-tools: Read Bash(echo:*) mda: requires: network: none --- # Compile basic Use this fixture to confirm target projection.README.mdupstream β
Contents are not inlined in the catalog. Open the upstream file.
Plain-text copy, no login: /skill-bundles/sno-ai-mda-conformance-compile-46-basic-targets-expected.md