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

93%

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.

An exceptionally dense, senior-level skill body: strict-ordered workflows, hard escalation gates, concrete commands and thresholds, and a pre-commit checklist with feedback loops. The main gaps are inline budget/billing detail that the skill itself admits lives in PR-2 CONTRACT.md, and under-signaled discovery of three of the four references/ bundle files.

Suggestions

Compress the Budget + Outbound Gating section to the two gate snippets and the CONTRACT.md pointer; move the plan-defaults table and MV stale cap detail to the referenced CONTRACT.md.

List each references/ file individually under 'Project invariants' (conversation-stages.md, humanization-rules.md, tool-patterns.md) with a one-line purpose so discovery does not depend on directory listing.

Reduce absolute dates to the minimum needed for ratchet context (e.g., keep the DeepSeek alias retirement date, drop 'hotfix 2026-05-01' provenance in §0).

Dimension	Reasoning	Score
Conciseness	Telegraphic and dense with zero padding; the glossary only defines project-specific terms (ratchet, Stranger Fig, turn). Not 5: the Budget/Outbound section inlines a full plan-defaults table and gating detail despite stating 'Detalle vivo en PR-2 CONTRACT.md', and several absolute dates ('post redesign 2026-04', 'hotfix 2026-05-01') add staleness risk — though the retired DeepSeek aliases are properly confined to the closed anti-patterns list.	4 / 5
Actionability	Copy-paste-ready commands ('find /home/chris/AISALESHT/backend/src -name "<basename>.py"', 'grep -rn "class <ClassName>" backend/src/shared/ backend/src/modules/'), executable snippets (budget_guard.check, outbound_rate_limiter.check), exact table names, thresholds (TIER_THRESHOLD = 200_000), and file paths cover the common cases.	5 / 5
Workflow Clarity	'Antes de codear (orden estricto)' is an explicitly ordered workflow with validation checkpoints ('Test reproductor antes del fix? RED → GREEN', 'Arch tests verde native'), a hard STOP/escalate gate for §3 surfaces, a 10-item pre-commit checklist, and explicit error-recovery guidance (traces-first diagnosis, /pm escalation).	5 / 5
Progressive Disclosure	Well-sectioned body with 'Project invariants (read on demand)' deferring references/sales-agent-brand-voice.md (a real bundle file) one level deep. Not 5: only one of the four references/ files is named individually — conversation-stages.md, humanization-rules.md, and tool-patterns.md are covered only by the generic 'references/ (pre-redesign conversation craft...)' pointer.	4 / 5
	Total	18 / 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 dense, highly specific description that explicitly states what it loads, when to use it, and a comprehensive list of natural trigger phrases. The only weaknesses are the persona framing ('Senior + arquitecto + CTO') and that it lists coverage areas rather than discrete actions.

Suggestions

Replace the 'Senior + arquitecto + CTO' persona clause with one more concrete capability statement (e.g., 'Guarda surfaces protegidas y anti-patterns del módulo sales_agent').

Trim the subsystem enumeration in the 'Use cuando' clause to the highest-signal items; the trigger list already covers discovery.

Dimension	Reasoning	Score
Specificity	Names the sales_agent module domain and several concrete loaded artifacts ('§3 protected surfaces, anti-patterns, decisiones cross-fase, checklist pre-commit') plus a rich subsystem list (voz de marca, scheduler/payment tools, observabilidad, callback handler, prompt cache, channel registry, closer studio, follow_up engine, eval loop, costo agentes). Not 5 because it enumerates scope more than actions and the 'Senior + arquitecto + CTO' persona framing is meta rather than a capability.	4 / 5
Completeness	Explicitly answers both what ('Carga §3 protected surfaces, anti-patterns, decisiones cross-fase, checklist pre-commit') and when ('Use cuando user pida cambio/feature/bug en sales_agent: ...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Extensive natural user phrasings with synonyms and variations: 'modifica sales_agent', 'bug en sales_agent', 'agregar tool al agente', 'agente no cierra', 'suena robótico', 'follow-up no dispara', 'el agente repite preguntas', 'PII en trazas', 'costo Kimi', 'eval goldens'. These are exactly what a user would say when needing this skill.	5 / 5
Distinctiveness Conflict Risk	Clear niche scoped to the sales_agent module post-2026-04 redesign; trigger terms are module-specific (closer studio, follow_up engine, costo Kimi, DeepSeek alias) with minimal overlap risk against generic 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/ap_sales_agent
Path: .claude/skills/sales-agent-expert/SKILL.md
Commit: d31f7bc

Reviewed: 16 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.