Browse Indicators¶

DAAS ships hundreds of prebuilt indicator rules across companies and countries, and the platform scales to thousands as you add datasources. This example shows how to browse them by entity, source, and op.

What is an "indicator"?¶

An indicator rule (row in indicator_rules) binds a source table + a math op + params to an indicator name, and its computed series lands in observations. For example, SPY_ma5 = sma over scraw_spy_daily.close with window=5.

Browse via SQL¶

From the repo root:

# How many indicator rules exist?
sqlite3 daas.db "SELECT count(*) FROM indicator_rules;"

# Indicators for SPY
sqlite3 daas.db "SELECT name, op, indicator_name, source_table FROM indicator_rules WHERE name LIKE 'SPY_%';"

# Distinct computed indicators in observations
sqlite3 daas.db "SELECT count(DISTINCT indicator) FROM observations;"

# Latest values for an indicator
sqlite3 daas.db "SELECT date, value FROM observations WHERE indicator='SPY_ma5' ORDER BY date DESC LIMIT 5;"

Browse by entity type¶

Entities are stocks or countries:

sqlite3 daas.db "SELECT entity_type, count(*) FROM entities GROUP BY entity_type;"
# stock|5575   country|60

So indicators span 5,575 stocks and 60 countries - company-level (SPY_ma5, AAPL_rsi14) and country/macro-level (CPI, GDP, treasury yields from cnstats / worldbank).

Browse via MCP¶

Point your MCP client at fd-daas-mcp and use the daas_* tools:

daas_list_indicators - all indicator rules (with effective scores).
daas_list_indicator_ops - the fixed math-op catalog + required params.
daas_get_series_latest - latest N rows of a series.
daas_calculate_indicator - ad-hoc compute over a source table without persisting (great for "what would this look like?").

Browse via skill¶

In Claude Code:

List all RSI indicators and show the latest value for each.

The fd-daas-fetch-data skill resolves the catalog and reads observations.

Available ops¶

uv run python .claude/skills/fd-daas-based-data-fetch/scripts/run_indicator.py --list-ops

Op	Params	Description
`sma` / `ema`	`window`	Simple/exponential moving average.
`rsi`	`window`	Relative strength index.
`pct_change` / `log_return` / `diff`	-	Change series.
`rolling_std` / `rolling_min` / `rolling_max`	`window`	Rolling stats.
`zscore`	`window`	Rolling z-score.
`ratio` / `level`	-	Ratio / level-of-series.

Next¶

Group the indicators you care about into a collection: Create Collections.