sf-research — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited sf-research (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
npm install -g @spfunctions/cli
# Optional: sf login (most read commands work without auth)The sf investigate and sf query commands work without authentication. sf me and account-scoped reads require sf login.
sf investigate is the read-only research router. It calls public world, query, econ, gov, traditional-market, and tool-planning surfaces in parallel, tolerates partial failures, and returns evidence + next commands in one envelope.
sf investigate "<topic>" --json --limit 5Example:
sf investigate "Fed rate cuts CPI" --json --limit 5This returns:
If the topic is broad or the user wants a different angle, also run:
sf query "<refined question>" --json --limit 3sf query is natural-language market search. Use it when you have a specific question, not a topic keyword.
For each top-3 market from Step 1, call:
sf inspect <ticker> --jsonReturns:
If the topic is macro or geopolitical, pull the world snapshot for the salience context:
sf world --json
# or for a specific topic
sf world --topic <topic-slug> --jsonThis shows what SF's salience ranker considers most material right now — divergences, residual volume signals, consensus breaks, resolving-soon markets.
Produce a 4-section research packet:
1. Market state What do the markets say about this topic? Cite the top-3 prices and what they imply.
2. Anchors What traditional markets / econ data corroborate or contradict the prediction-market read? (e.g., "WTI at $100 is consistent with the Iran-risk market at 45¢")
3. Catalysts What upcoming events would resolve or move the market? Pull from sf investigate output's catalyst field.
4. Next actions Concrete next commands: sf inspect <ticker> for a specific market, sf watchlist add to track, sf create to build a thesis.
If the user wants to turn the research into a position:
sf create "<thesis statement based on research>"See the sf-thesis-build skill for the full thesis lifecycle.
Each --json command returns an envelope:
{
"status": "ok" | "partial" | "error",
"data": { ... },
"warnings": [],
"nextActions": [ { "description": "...", "command": "..." } ]
}Always check status before assuming data is complete.
world.read is cost-bearing. Set SF_API_KEY via sf login, or use the public GET /api/agent/world endpoint (free).warnings array. Other data is still usable.sf scan --keyword "<topic>" --json for broader search.sf-discover — find quality-ranked opportunities (not topic-specific research)sf-inspect — deep dossier on one tickersf-thesis-build — turn research into a structured causal tree~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.