tier3-probes

Catalogo de feasibility probes para champions Tier 3 — Scrapling, Oumi, Memvid, BERTopic, Reranker, PDF extract

Best use case

tier3-probes is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Catalogo de feasibility probes para champions Tier 3 — Scrapling, Oumi, Memvid, BERTopic, Reranker, PDF extract

Teams using tier3-probes should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/tier3-probes/SKILL.md --create-dirs "https://raw.githubusercontent.com/gonzalezpazmonica/pm-workspace/main/.claude/skills/tier3-probes/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/tier3-probes/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How tier3-probes Compares

Feature / Agenttier3-probesStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Catalogo de feasibility probes para champions Tier 3 — Scrapling, Oumi, Memvid, BERTopic, Reranker, PDF extract

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Skill: Tier 3 Probes

> Feasibility probes antes de instalar dependencias pesadas.
> Ref: Era 183 Tier 3 Champions.

## Cuando usar

- Antes de `pip install` pesado para un champion Tier 3
- Verificar entorno tras travel-pack a maquina nueva
- Auditoria periodica de preconditions ML stack
- Pre-activacion de MCP templates (scrapling opt-in)

## Inventario

| Probe | Champion | Spec | Verifica |
|---|---|---|---|
| `scrapling-probe.sh` | Scrapling | SE-061 | Python >= 3.10, scrapling, lxml, opcional playwright+chromium |
| `oumi-probe.sh` | Oumi training | SE-028 | Python 3.10+, oumi, torch, disk >= 10GB |
| `memvid-probe.sh` | Memvid backup | SE-041 | Python 3.10+, memvid |
| `bertopic-probe.sh` | BERTopic | SE-033 | Python, bertopic, sentence-transformers, UMAP, HDBSCAN |
| `reranker-probe.sh` | Cross-encoder | SE-032 | Python, sentence-transformers, torch, BAAI model availability |
| `pdf-extract-probe.sh` | PDF chain | SPEC-102/103/104 | Python, pdfminer, java deps |

## Invocacion uniforme

Todos los probes siguen el mismo contrato:

```bash
bash scripts/X-probe.sh            # Verbose output
bash scripts/X-probe.sh --json     # Machine-readable
```

### Output schema (JSON)

```json
{
  "verdict": "VIABLE|NEEDS_INSTALL|BLOCKED",
  "python_version": "3.12.3",
  "{package}_installed": 0|1,
  "disk_free_gb": 100,
  "reasons": ["human-readable explanation"]
}
```

### Exit codes

- `0` — VIABLE o NEEDS_INSTALL (no bloqueador)
- `1` — BLOCKED (Python incompatible, disk insuficiente, dep critica ausente)
- `2` — usage error (flag desconocido)

## Casos de uso

### Pre-install check
```bash
bash scripts/scrapling-probe.sh --json | jq .verdict
# → "VIABLE" → proceder con `pip install scrapling`
```

### Batch check environment readiness
```bash
for p in scrapling oumi memvid bertopic reranker pdf-extract; do
  echo "=== $p ==="
  bash scripts/$p-probe.sh --json | jq .verdict
done
```

### CI gate (opcional)
Un PR que añade champion Tier 3 puede verificar que el probe devuelve VIABLE|NEEDS_INSTALL antes de merge.

## No hacen

- No instalan dependencias (solo verifican)
- No descargan modelos
- No ejecutan inference (eso es Slice 2+ de cada champion)
- No auto-fixer (requiere intervención humana)

## Patron de diseno (SE-061 reference)

Cada probe:
1. Verifica herramientas nativas (python3, pip3, df)
2. Detecta deps Python via `python3 -c "import X" 2>/dev/null`
3. Mide disk free
4. Clasifica verdict por 3 categorias
5. Lista reasons humanas para cada fallo
6. JSON estable para automatizacion

Replicar este patron para nuevos Slice 1 de champions futuros.

## Referencias

- SE-061 Scrapling: `docs/propuestas/SE-061-scrapling-research-backend.md`
- SE-032 Reranker: `docs/propuestas/SE-032-reranker-layer.md`
- SE-033 BERTopic: `docs/propuestas/SE-033-topic-cluster-skill.md`
- SE-041 Memvid: `docs/propuestas/SE-041-memvid-portable-memory.md`
- SE-028 Oumi: `docs/propuestas/SE-028-oumi-training-pipeline.md` (diferido)
- Roadmap Era 183: `docs/ROADMAP.md` §Tier 3 Champions

Related Skills

zoom-out

32
from gonzalezpazmonica/pm-workspace

Elevates perspective from trees to forest. Maps architecture, dependencies, and second-order effects before implementation decisions. Use when designing, when evaluating trade-offs, or at the start of design sessions.

workspace-integrity

32
from gonzalezpazmonica/pm-workspace

Catalogo de integrity auditors — drift CLAUDE.md, rule manifest, orphan rules, agents catalog sync, baseline, agent size

wellbeing-guardian

32
from gonzalezpazmonica/pm-workspace

Sistema proactivo de bienestar individual

web-research

32
from gonzalezpazmonica/pm-workspace

Search the web to resolve context gaps — documentation, versions, CVEs, best practices. Auto-starts SearxNG Docker if available, falls back to WebSearch.

voice-inbox

32
from gonzalezpazmonica/pm-workspace

Transcripción de audio y flujo audio→texto→acción para mensajes de voz

verification-lattice

32
from gonzalezpazmonica/pm-workspace

Multi-layer verification pipeline beyond Code Review

topic-cluster

32
from gonzalezpazmonica/pm-workspace

BERTopic clustering — agrupa retros/PBIs/incidents/lessons en topics tematicos con labels. Filtra ruido, descubre patrones cross-proyecto

time-tracking-report

32
from gonzalezpazmonica/pm-workspace

Generación de informes de imputación de horas a Excel/Word

test-architect

32
from gonzalezpazmonica/pm-workspace

Design and generate highest-quality tests across 16 languages and 14 test types

tech-research-agent

32
from gonzalezpazmonica/pm-workspace

Agente de investigación técnica autónoma — investiga temas, genera informes, notifica al humano designado

team-onboarding

32
from gonzalezpazmonica/pm-workspace

Onboarding y evaluación de competencias para nuevos miembros del equipo

team-coordination

32
from gonzalezpazmonica/pm-workspace

Multi-team orchestration — create teams, assign members, detect cross-team blockers