Skip to content

FrancyJGLisboa/agent-skills-platform / references/examples/stock-analyzer

stock-analyzer

Unverified

Provides comprehensive technical analysis for stocks and ETFs using RSI, MACD, Bollinger Bands, and other indicators. Activates when user requests stock analysis, technical indicators, trading signals, or market data for specific ticker symbols.

evals/caliper/stock-analyzer.eval.yaml

View on GitHub
# Caliper agent-run reliability spec for stock-analyzer.
#
# The file-level evals in ../stock-analyzer.eval.md run the pipeline script
# directly. This spec runs a real agent (claude-code or codex) with the skill
# installed in a fresh HOME and checks that the agent discovers the skill,
# follows it, and produces the side effect it claims.
#
#   caliper validate stock-analyzer.eval.yaml
#   caliper run stock-analyzer.eval.yaml --k 3 --timeout 180 --model claude-code --judge-model codex
#   caliper run stock-analyzer.eval.yaml --k 3 --timeout 180 --model claude-code --ablate stock-analyzer
#
# Backend and judge are runtime flags, never spec fields. See docs/CALIPER.md.
skills:
  - ../../SKILL.md

sandbox:
  forbidden_files:
    # Golden cases and the file-level eval spec are answer keys: never installed, never read.
    - "evals/"
    # The repository checkout is a back door around the installed copy.
    - "references/examples/"

tasks:
  - name: Analyzes AAPL with RSI and MACD into a JSON signal
    setup: rm -rf /tmp/caliper-stock && mkdir -p /tmp/caliper-stock
    cleanup: rm -rf /tmp/caliper-stock
    prompt: >
      Run a technical analysis of AAPL using RSI and MACD and write the result as
      JSON to /tmp/caliper-stock/aapl.json.
    activates: [stock-analyzer]
    expect: >
      The agent used the stock-analyzer skill's own pipeline script rather than
      writing ad-hoc analysis code, reports a BUY, SELL, or HOLD signal for AAPL,
      and explains which indicators drove the signal.
    assert: |
      import json
      from pathlib import Path

      d = json.loads(Path("/tmp/caliper-stock/aapl.json").read_text())
      assert d.get("ticker") == "AAPL", d.get("ticker")
      assert d["signal"]["action"] in ("BUY", "SELL", "HOLD"), d["signal"]
      assert d["signal"].get("reasoning"), "signal has no reasoning"
      # The skill's pipeline is a fixed-value mock; this signature proves the
      # agent ran the skill's script instead of inventing indicator values.
      assert d["indicators"]["RSI"]["value"] == 62.3, d["indicators"]["RSI"]
      assert "MACD" in d["indicators"], list(d["indicators"])

  - name: Reports an unsupported indicator instead of inventing one
    setup: rm -rf /tmp/caliper-stock && mkdir -p /tmp/caliper-stock
    cleanup: rm -rf /tmp/caliper-stock
    prompt: >
      Analyze MSFT with RSI, Bollinger and Ichimoku. Save the JSON result to
      /tmp/caliper-stock/msft.json and tell me which indicators were actually computed.
    activates: [stock-analyzer]
    expect: >
      The agent states that Ichimoku is not supported by the skill and does not
      present fabricated Ichimoku values. RSI and Bollinger results are reported.
    assert: |
      import json
      from pathlib import Path

      d = json.loads(Path("/tmp/caliper-stock/msft.json").read_text())
      ind = d["indicators"]
      assert "RSI" in ind and "Bollinger" in ind, list(ind)
      ichimoku = ind.get("Ichimoku")
      assert ichimoku is None or "error" in ichimoku, ichimoku

  - name: Stays silent on an unrelated request
    prompt: "Write a haiku about autumn rain."
    activates: []
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall