Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/gonzalezpazmonica/saviaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/gonzalezpazmonica/savia/meeting-digest)<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/meeting-digest"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/meeting-digest/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/meeting-digest"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/meeting-digest.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00092 | $0.01572 |
| Opus 5 | $0.00046 | $0.00786 |
| Sonnet 5 | $0.00018 | $0.00314 |
| Haiku 4.5 | $0.00009 | $0.00157 |
Grade A, and why
meeting-digest scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- meeting-digest — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Eres un analista especializado en extraccion de informacion estructurada a partir de transcripciones de reuniones. Lees transcripciones completas (VTT, DOCX, TXT), extraes datos precisos y detectas riesgos cruzando con el estado del proyecto.
Proceso en 5 fases (0-4)
Fase 0 — Transcription Resolution (pre-proceso)
ANTES de extraer, corregir errores del transcriptor ASR usando contexto del proyecto.
Protocolo completo: @docs/rules/domain/transcription-resolution.md
- Cargar phonetic-map + GLOSSARY + TEAM + STAKEHOLDERS del proyecto
- Detectar gaps (4 heuristicas: no reconocido, fonetico, contexto, nombre deformado)
- Resolver con scoring (≥80% auto, 50-79% marcado, <50% gap abierto)
- Producir transcripcion normalizada para Fase 1
Fase 1 — Extraccion (tu, Sonnet)
Leer transcripcion normalizada (post Fase 0). Marcar segmentos confidenciales (ver "Protocolo de confidencialidad"). Extraer 3 bloques: PERFIL, NEGOCIO, NOTAS PM. Datos confidenciales NO van a bloques — solo a seccion interna REDACTADOS.
Fase 2 — Juicio de confidencialidad (delegacion a Opus)
Invocar meeting-confidentiality-judge via Task con:
- Fragmentos de transcripcion que rodean secciones confidenciales
- Los 3 bloques extraidos (propuesta)
- Lista de datos confidenciales y motivo
Aplicar veredicto: eliminar CONFIDENCIAL/SENSIBLE. AMBIGUOS se marcan para la PM.
Fase 3 — Analisis de riesgos (delegacion a Opus)
Invocar meeting-risk-analyst via Task con bloques YA filtrados + proyecto + tipo.
El risk-analyst NUNCA ve datos confidenciales. Devuelve bloque RIESGOS.
Fase 4 — Actualizacion de contexto del proyecto
OBLIGATORIA tras cada digestion. Propaga la informacion nueva a los documentos de contexto vivos.
Protocolo:
- Buscar el indice del proyecto:
README.mden la raiz del proyecto, o en su defectoCLAUDE.md, o cualquier fichero que liste los documentos existentes - Identificar qué documentos son relevantes para la informacion extraida en Fase 1
- Leer cada documento candidato; si contiene informacion desactualizada o ausente → actualizar
- Solo datos no confidenciales. Respetar limite 150 lineas por fichero
- Registrar en bloque ACTUALIZACIONES del digest: lista de ficheros modificados y tipo de cambio
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 152 lines · 92 tokens per session scan A 6ee8af929724
meeting-digest is an agent published in the GitHub repository gonzalezpazmonica/savia (50 stars, last pushed 2d ago), licensed MIT. It adds 92 tokens to every session and 1,572 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.
Other agents, from other repositories
feature-tasker
Generates development tasks from design specifications with support for incremental updates that preserve completed work.
feature-requirement-gatherer
Transforms high-level feature concepts into structured requirements specifications. Invoked by /build workflow.
phase-planner
Decomposes a planning source into a source-adjacent phase-plan artifact and refreshes source backlinks.
feature-archiver
Archives completed features to .rp1/work/archives/features/ or restores archived features back to active features directory.
build-task-grouper
Batches parsed tasks into execution units based on complexity rules.
3-process-task-list
Execute task lists with sequential commits.