WaggleKiller Bee packs for Buzz

risk-portfolio

Value at Risk with real backtests, portfolio construction beyond Markowitz, factor attribution, credit risk and the copulas that tie the tails together.

version
0.1.0
author
Saulo Duttra
license
Apache-2.0
providers
unset
verified against
buzz @ ed4b3e7a

Teams

Risk & Portfolio

Value at Risk with real backtests, portfolio construction beyond Markowitz, factor attribution, credit risk and the copulas that tie the tails together.

Team instructions, in full

You are one specialist among several on the same desk. Answer from your own area and say plainly when a question belongs to someone else's — naming which. Every quantitative claim carries its formula, its assumptions, and the regime where it stops holding. When another member's answer contradicts yours, say so explicitly rather than softening it. None of you gives investment advice.

var-labport-labfactor-labcredit-labcopula-lab

risk-portfolio.team.jsonrisk-portfolio.team.png
sha256 checksums
risk-portfolio.team.json 18,708 B
af07808b150579c52c820bc09aa76c127b6eeaf524c3f7f3ef56ddf33a988a58
risk-portfolio.team.png 50,870 B
24ce8b9c839e59e23ff6cf2aae5ac1d7ec361d00c007a5980325131734ac16eb
Post as a chat card

Paste the link as the message body and the imeta tag as its media tag. Buzz renders it as an importable agent card instead of a file attachment — the x value is the same sha256 published above, and the card refuses to offer Import without it.

risk-portfolio.team.json

[risk-portfolio.team.json](https://killer-bee-4rn.pages.dev/downloads/risk-portfolio/risk-portfolio.team.json)
["imeta","url https://killer-bee-4rn.pages.dev/downloads/risk-portfolio/risk-portfolio.team.json","m application/json","x af07808b150579c52c820bc09aa76c127b6eeaf524c3f7f3ef56ddf33a988a58","size 18708","filename risk-portfolio.team.json"]

risk-portfolio.team.png

[risk-portfolio.team.png](https://killer-bee-4rn.pages.dev/downloads/risk-portfolio/risk-portfolio.team.png)
["imeta","url https://killer-bee-4rn.pages.dev/downloads/risk-portfolio/risk-portfolio.team.png","m image/png","x 24ce8b9c839e59e23ff6cf2aae5ac1d7ec361d00c007a5980325131734ac16eb","size 50870","filename risk-portfolio.team.png"]

Personas

  • Value at Risk

    Computes VaR three ways (historical, parametric, Monte Carlo) plus Expected Shortfall two ways and backtests the result with Kupiec POF and Christoffersen independence.

    no model set

    recruitment2 / 32 parallel

    Compiles to the agent's native parallelismfield. The 1–32 range is Buzz's, not ours.

    thresholdlow

    How little it takes to get a response. In the desktop import, low and medium compile to respondTo: anyone (mention-triggered, like every imported agent). Low additionally writes require_mention = false into the pack's ACP rules file — which only applies if you run buzz-acp yourself with --subscribe config.

    persistencemedium

    How long it stays on a task. Compiles to idle and turn timeouts.

    propagationhigh

    Catalog metadata only. Nothing at runtime reads this — it says how freely the author expects the pack to be forked.

    var-lab.agent.jsonvar-lab.agent.pngRead the prompt
    sha256 checksums
    var-lab.agent.json 3,235 B
    f0c262f476174f9d030a4dc9b072915fcde3b1175e9783d4ac3043a3d2f4d07c
    var-lab.agent.png 27,703 B
    46dd0e5d8da8a8b2be5b24702ef1eecb2623cb5631ffc48c5795b53110b6429f

  • Allocation

    Turns expected returns and a covariance matrix into portfolio weights via Markowitz, Black-Litterman, Risk Parity, or HRP.

    no model set

    recruitment4 / 32 parallel

    Compiles to the agent's native parallelismfield. The 1–32 range is Buzz's, not ours.

    thresholdmedium

    How little it takes to get a response. In the desktop import, low and medium compile to respondTo: anyone (mention-triggered, like every imported agent). Low additionally writes require_mention = false into the pack's ACP rules file — which only applies if you run buzz-acp yourself with --subscribe config.

    persistencemedium

    How long it stays on a task. Compiles to idle and turn timeouts.

    propagationhigh

    Catalog metadata only. Nothing at runtime reads this — it says how freely the author expects the pack to be forked.

    port-lab.agent.jsonport-lab.agent.pngRead the prompt
    sha256 checksums
    port-lab.agent.json 3,373 B
    b3a04f6d3ebce361a04fb55f0e5c57701a6d081e089a3b51f049dc245d0f4433
    port-lab.agent.png 27,757 B
    d9524fdd436e6b168ead97dab8c9ec9ca487f6284d198810bd70622839469baa

  • Factor Models

    Builds cross-sectional factor models, Fama-MacBeth premia, pure-factor portfolios, and Euler risk attribution from raw matrix algebra.

    no model set

    recruitment8 / 32 parallel

    Compiles to the agent's native parallelismfield. The 1–32 range is Buzz's, not ours.

    thresholdmedium

    How little it takes to get a response. In the desktop import, low and medium compile to respondTo: anyone (mention-triggered, like every imported agent). Low additionally writes require_mention = false into the pack's ACP rules file — which only applies if you run buzz-acp yourself with --subscribe config.

    persistencelong

    How long it stays on a task. Compiles to idle and turn timeouts.

    propagationmedium

    Catalog metadata only. Nothing at runtime reads this — it says how freely the author expects the pack to be forked.

    factor-lab.agent.jsonfactor-lab.agent.pngRead the prompt
    sha256 checksums
    factor-lab.agent.json 3,575 B
    46be9d987a61f44fefbd4112d328ee4aba1032d0476c1c8ef86aa1ef73bb8842
    factor-lab.agent.png 28,496 B
    06bd1115f9afaf31ed6fd5524aaed7570bb9f009513115cf11dffeda2e1a253f

  • Credit Risk

    Prices default risk end to end: Merton structural model, hazard curves, CDS legs and bootstrap, and defaultable bonds.

    no model set

    recruitment2 / 32 parallel

    Compiles to the agent's native parallelismfield. The 1–32 range is Buzz's, not ours.

    thresholdmedium

    How little it takes to get a response. In the desktop import, low and medium compile to respondTo: anyone (mention-triggered, like every imported agent). Low additionally writes require_mention = false into the pack's ACP rules file — which only applies if you run buzz-acp yourself with --subscribe config.

    persistencelong

    How long it stays on a task. Compiles to idle and turn timeouts.

    propagationmedium

    Catalog metadata only. Nothing at runtime reads this — it says how freely the author expects the pack to be forked.

    credit-lab.agent.jsoncredit-lab.agent.pngRead the prompt
    sha256 checksums
    credit-lab.agent.json 3,660 B
    4d5462dd8ea1721c5f1c32ff8b71eede3b3b818e6628c8e386fe837e038506f0
    credit-lab.agent.png 28,051 B
    517eeade69ee73d306d9e5570a5ea46c86c120833f5f832d70d7ef8f311cc38e

  • Copulas

    Models dependence separately from margins: five bivariate copula families with exact samplers, tail-dependence coefficients, and tau-inversion or MLE fitting.

    no model set

    recruitment5 / 32 parallel

    Compiles to the agent's native parallelismfield. The 1–32 range is Buzz's, not ours.

    thresholdmedium

    How little it takes to get a response. In the desktop import, low and medium compile to respondTo: anyone (mention-triggered, like every imported agent). Low additionally writes require_mention = false into the pack's ACP rules file — which only applies if you run buzz-acp yourself with --subscribe config.

    persistencelong

    How long it stays on a task. Compiles to idle and turn timeouts.

    propagationmedium

    Catalog metadata only. Nothing at runtime reads this — it says how freely the author expects the pack to be forked.

    copula-lab.agent.jsoncopula-lab.agent.pngRead the prompt
    sha256 checksums
    copula-lab.agent.json 3,730 B
    16a21e3970fbd0373daaae87205a6ea4e6da637af3f8857686ac7b5e71b6e802
    copula-lab.agent.png 28,459 B
    b8fbd7de91c0a1f9b13d35abf2fd617fcc9544bf15b9765b7a1fbbc7eccec69f

Installing

4 clicks plus the OS file picker. There is no drag-and-drop shortcut: the Agents section has no drop target.

  1. Download the file below.
  2. In Buzz Desktop, open the Agents section in the sidebar.
  3. Click the "+" card, then "Import".
  4. Pick the file, review the preview, and confirm.

Imported is not running

  • The agent exists but is not running yet.
  • It needs provider credentials from the app's global settings.
  • Adding it to a channel is a separate action in the agent's profile panel.

Or build it from source

The files above are generated from this repository. Nothing is fetched at runtime and nothing phones home.

uv run python -m killerbee build packs/risk-portfolio