sales-agent-expert

Senior + arquitecto + CTO del módulo sales_agent post redesign 2026-04. Carga §3 protected surfaces, anti-patterns, decisiones cross-fase, checklist pre-commit. NO carga code vivo (paths/LOC/tests cambian). Use cuando user pida cambio/feature/bug en sales_agent: voz de marca, scheduler/payment tools, observabilidad, callback handler, prompt cache, channel registry, semantic router, closer studio, follow_up engine, eval loop, costo agentes. Triggers: 'modifica sales_agent', 'bug en sales_agent', 'agregar tool al agente', 'agente no cierra', 'suena robótico', 'voz de marca', 'agregar canal sales', 'modificar prompt del specialist', 'closer studio', 'follow-up no dispara', 'el agente repite preguntas', 'PII en trazas', 'costo Kimi', 'DeepSeek alias', 'tier pricing 200k', 'eval goldens'.

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

77%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 dense, expert-grade context loader with strong workflow sequencing and validation checkpoints, but its progressive disclosure is weak: most bundle files are unlinked from the body and substantial reference-grade detail is inlined rather than split out.

Suggestions

Progressive disclosure: add one-line signals in the body for the three unreferenced bundle files (conversation-stages.md, humanization-rules.md, tool-patterns.md), or explicitly mark them deprecated so the bundle structure is navigable.

Progressive disclosure: move the inlined "Surfaces compartidas con copilot" enumeration and the Budget+Outbound Gating detail into a reference file, keeping only the anchor and the §3 gate summary in SKILL.md.

Conciseness: collapse the dated "Decisiones cardinales últimos 60 días" changelog into a pointer to docs/process/learnings.md, retaining only the few decisions that change current behavior and dropping commit hashes.

Dimension	Reasoning	Score
Conciseness	Extremely dense and free of basic-concept padding, assuming Claude's competence throughout; however the dated "Decisiones cardinales últimos 60 días" changelog (commit hashes, specific dates) and minor repetition of anti-duplication across §0 and the shared-surfaces section could be trimmed.	4 / 5
Actionability	Provides concrete executable commands ("cat .claude/rules/anti-duplication.md", "find $(git rev-parse --show-toplevel)/backend/src -name ...", "grep -rn ...") and real code signatures (BudgetGuard.check, OutboundRateLimiter.check), but much of the body is decisional/anti-pattern context rather than copy-paste steps, leaving minor gaps.	4 / 5
Workflow Clarity	Clear strict-ordered "Antes de codear" sequence, a 3-step escalation workflow in §0 with STOP/escalate checkpoints, a RED→GREEN test feedback loop, and a 10-item pre-commit checklist — explicit validation and error-recovery loops throughout.	5 / 5
Progressive Disclosure	Sections are well organized and a Pointers block exists, but only 1 of the 4 bundle files (references/sales-agent-brand-voice.md) is signaled from the body, and large detail blocks (the 13-item shared-surfaces enumeration, the full Budget+Outbound Gating section) are inlined in SKILL.md rather than delegated to reference files.	3 / 5
	Total	16 / 20 Passed

Description

95%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 strong, highly specific description that explicitly covers what the skill loads, what it deliberately avoids, and when to invoke it, backed by a rich set of natural trigger phrases. The only minor weakness is the "loads context" framing rather than discrete action verbs.

Dimension	Reasoning	Score
Specificity	Lists many concrete surfaces and capabilities ("§3 protected surfaces, anti-patterns, decisiones cross-fase, checklist pre-commit", "voz de marca, scheduler/payment tools, observabilidad, callback handler") with comprehensive coverage, but the framing is "loads/uses" context rather than discrete executable actions, leaving a minor gap versus the top anchor.	4 / 5
Completeness	Clearly answers both "what" ("Carga...", "NO carga code vivo") and "when" ("Use cuando user pida cambio/feature/bug en sales_agent") with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	An explicit "Triggers:" block supplies natural user phrases ("'modifica sales_agent'", "'agente no cierra'", "'suena robótico'", "'PII en trazas'", "'costo Kimi'") with comprehensive coverage including synonyms and variations a user would actually say.	5 / 5
Distinctiveness Conflict Risk	Tightly scoped to the "sales_agent" module niche with distinct, module-specific triggers, giving minimal overlap risk with other skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: alpacapurpura/luana-method
Path: claude/skills/sales-agent-expert/SKILL.md
Commit: caa08df

Reviewed: ~22 hours 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.