graph-run-discipline

Disciplina de ejecucion de un graph aprobado: gate de aprobacion con graph_id+version exactos, ejecucion por olas topologicas con enforcement de write_set, receipts observables con paths_written, fallos tipados (12 codigos), integracion sin outputs sueltos y loop maker/checker con stop conditions. Usar al pedir 'ejecuta el graph aprobado', 'olas de ejecucion', 'enforcement de write set', 'receipts de nodos'. NO para compilar el DAG (graph-spec-compiler) ni coordinar agent teams (agent-team-orchestration).

Quality

86%

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

83%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.

The body is a well-structured, concise overview with a sequenced validated procedure and real one-level-deep references. It is actionable and assumes competence, with only minor tightening and inline-signaling opportunities.

Suggestions

Make the integration step (step 5) more executable by naming the concrete artifact/manifest path and a runnable comparison command, mirroring the executable style of the pre-vuelo step.

In the Packet block, signal the three concrete reference files inline (e.g. '**Gate**: See references/gate-de-aprobacion.md') instead of only describing the references/ layer generically, so navigation is one-click.

The failure-codes asset path resolves to the sibling graph-spec-compiler skill; either surface a local copy or a short enumerated list of the 12 codes inline so the skill is self-contained for the classification step.

Dimension	Reasoning	Score
Conciseness	Lean prose that assumes Claude's competence and avoids explaining what a graph or a receipt is; only minor trims possible (e.g. the Packet prose could be slightly tightened), so it sits above 3 but not perfectly lean at 5.	4 / 5
Actionability	Provides an exact executable command (python3 .../validate_graph_spec.py <spec> --json) and concrete rules (APROBAR <graph_id> <version>, max 2 retries, stop conditions), with minor gaps such as the integration and wave-execution steps being described rather than given as runnable commands.	4 / 5
Workflow Clarity	The 6-step Procedimiento is explicitly sequenced with validation checkpoints (cero hallazgos o no hay corrida, paths_written vs write_set comparison, max 2 ciclos, stop conditions), matching the anchor for clear sequence with explicit validation and feedback loops.	5 / 5
Progressive Disclosure	Body is a concise overview pointing one level deep to real reference files (gate-de-aprobacion.md, ejecucion-por-olas.md, maker-checker.md), all of which exist in references/, plus scripts/check.sh and assets/manifest.json; structure is good but the Packet block enumerates layer categories generically rather than signaling each concrete reference inline, leaving minor organization gaps versus a fully clean 5.	4 / 5
	Total	17 / 20 Passed

Description

90%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.

The description is strong across all dimensions: specific, trigger-rich, complete with both what-and-when, and clearly bounded against sibling skills. It reads in third person and avoids fluff.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions (gate de aprobacion, ejecucion por olas, enforcement de write_set, receipts con paths_written, fallos tipados de 12 codigos, loop maker/checker) with only minor coverage gaps around the actual integration step detail.	4 / 5
Completeness	Explicitly answers both 'what' (disciplina de ejecucion de un graph aprobado with enumerated capabilities) and 'when' (concrete 'Usar al pedir...' triggers plus an explicit 'NO para...' boundary), matching the anchor for clear what-and-when with trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases quoted verbatim ('ejecuta el graph aprobado', 'olas de ejecucion', 'enforcement de write set', 'receipts de nodos') plus metadata triggers, covering the synonyms a user would actually say.	5 / 5
Distinctiveness Conflict Risk	Carves a clear niche (runtime execution discipline of an approved graph) and explicitly names what it is NOT for (graph-spec-compiler compilation, agent-team-orchestration), minimizing overlap and 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-run-discipline/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.