Saltar al contenido
TakoAPI
Volver a skills

babel-epistemic

por mdiskint

Coding Agents & IDEs
421Descargas según clawskills.sh
0Vistas
0 me gustaInicia sesión para dar me gusta a esta skill
—Aún sin valoraciones
Tu valoración

Resumen

Prevents metacognitive poisoning in multi-agent handoffs.

Descripción

Qué hace esta skill

Babel Epistemic Convention tracks confidence levels across multi-agent handoffs. Without it, inferences made by one agent silently become verified facts by the time they reach a third agent. The skill provides a tagging vocabulary for confidence, intent, register, affect, grounds, and trajectory that travels with the work.

Confidence calibration propagates through the forwarding block without any shared infrastructure, so the receiving agent does not need the skill installed.

Cuándo usarla

  • Tagging a news-sourced inference as REPORTED before passing it to a downstream summarizer
  • Preventing a derived metric from being treated as confirmed data in a financial pipeline
  • Flagging regulatory constraints that must remain non-overridable across agent delegations
  • Marking a four-month declining trend as ESCALATE rather than a neutral INFORM update
  • Auditing a three-agent research chain for confidence laundering after incorrect outputs

Flujo de trabajo de ejemplo

ENTRADA

User asks: Research Acme Corp's potential Vanta partnership and hand off to the drafting agent.

AGENTE
  1. 1

    Search for partnership mentions and find a single unverified news item

  2. 2

    Label the claim REPORTED at 0.25 confidence, not VERIFIED_DATA

  3. 3

    Set intent to INFORM and consider switching to FLAG_RISK given low confidence

  4. 4

    Append the Babel forwarding block to the handoff output

  5. 5

    Drafting agent inherits the label as REPORTED and does not promote it to verified in the draft

SALIDA

Handoff payload with per-assertion confidence tags intact; drafting agent qualifies the claim as unconfirmed in the output.

Instalación

clawhub install mdiskint/babel-epistemic

Ejecuta el comando de la CLI o pega el prompt en Claude Code, Cursor, Codex o el agente de programación que uses.

Enlaces