graph-spec-compiler

Compila un objetivo en lenguaje natural a un DAG validado: objective_contract, nodos con write_set declarado, edges con payload nombrado, olas topologicas, guardian de contexto fresco con manifiesto cerrado y controles falsificables — con validador determinista y paquete de aprobacion APROBAR/REVISAR/CANCELAR. Usar al pedir 'compila mi objetivo en un graph', 'DAG con olas paralelas', 'write sets disjuntos', 'graph engineering'. NO para diagramas mermaid ni para workflows JavaScript (dynamic-workflow-forge).

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A concise, well-structured body with an executable, validation-gated workflow and real referenced files. The main weakness is progressive disclosure: several referenced paths and Packet-listed directories do not exist in the bundle.

Suggestions

Add the missing templates/ directory (graph-spec.skeleton.json and approval-package.md) referenced in the Procedimiento, or remove the references if those artifacts are generated elsewhere.

Reconcile the Packet block's listed layers (knowledge/, prompts/, examples/, agents/) with the actual bundle — either add the directories or trim the list to assets/, references/, scripts/ that actually ship.

Inline a minimal copy-paste graph_spec example (or a pointer to scripts/fixtures/valid-graph.json) so the validate command can be run immediately without searching.

Dimension	Reasoning	Score
Conciseness	Lean and efficient: assumes Claude's competence, never explains what a DAG or graph is, and every section earns its place (e.g. 'Convierte una peticion simple en un graph_spec verificable ... El validador scripts/validate_graph_spec.py es determinista'). Functional provenance tags like [CÓDIGO]/[DOC] add only minimal overhead.	5 / 5
Actionability	Provides concrete executable guidance — 'python3 scripts/validate_graph_spec.py <spec.json> --json', '--emit-fix', 'APROBAR <graph_id> <version>' — with real file paths, but no full copy-paste spec example is inlined inline (one exists only in scripts/fixtures/), leaving a minor gap.	4 / 5
Workflow Clarity	A clear 5-step procedure with an explicit validation checkpoint ('Valida ... cero hallazgos o corrige con --emit-fix') and a hard stop/approval gate ('DETENTE: nada se ejecuta sin APROBAR <graph_id> <version>'), including a fix-and-retry feedback loop for an approval-gated operation.	5 / 5
Progressive Disclosure	Well-organized with one-level-deep, clearly signaled references that exist (references/disciplina-de-escritura.md, references/fases-de-compilacion.md, assets/graph-spec.schema.json, scripts/validate_graph_spec.py), but the body and Packet block also reference templates/graph-spec.skeleton.json, templates/approval-package.md and list knowledge/, prompts/, examples/, agents/, templates/ directories that are not present in the bundle.	4 / 5
	Total	18 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A highly specific, complete description that states concrete capabilities, explicit positive triggers, and a negative boundary. Only minor weakness is that some trigger terms lean toward specialized jargon rather than everyday user phrasing.

Dimension	Reasoning	Score
Specificity	Names the domain and lists multiple concrete actions — 'objective_contract, nodos con write_set declarado, edges con payload nombrado, olas topologicas, guardian de contexto fresco ... controles falsificables ... validador determinista ... paquete de aprobacion APROBAR/REVISAR/CANCELAR' — comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both 'what' (compiles NL into a validated DAG with the listed components and approval package) and 'when' ('Usar al pedir ...') with concrete trigger phrases, plus a negative boundary ('NO para diagramas mermaid ni ... dynamic-workflow-forge').	5 / 5
Trigger Term Quality	Includes natural quoted trigger phrases ('compila mi objetivo en un graph', 'DAG con olas paralelas', 'write sets disjuntos', 'graph engineering') with good coverage, but several are niche jargon a user may not spontaneously say rather than everyday synonyms; a few natural terms are missing.	4 / 5
Distinctiveness Conflict Risk	Clear niche (graph-spec compilation vs. runtime execution and control design) with an explicit boundary against mermaid-diagramming and dynamic-workflow-forge, giving minimal conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

68%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 11 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	11 / 16 Passed

Repository: JaviMontano/claude-plugins
Path: plugins/claude-native-toolkit/skills/graph-spec-compiler/SKILL.md
Commit: e8f986b

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.