Screen companies by signals across all source types — filings, earnings, transcripts, IR events. This is NOT metric screening (use screen_companies for P/E, ROIC, etc.). Screens by verifiable facts, not LLM-generated scores. Available signals: Filing intelligence (from 10-K/10-Q): - tone_cautious...
Risk signalsHigh parameter count (10 properties)
Part of the MetricDuck — Financial Analysis server.
Free to start. No card required.
AI agents invoke screen_filing_signals to trigger processes or run actions in MetricDuck — Financial Analysis. Execute operations can have side effects beyond the immediate call -- triggering builds, sending notifications, or starting workflows. Rate limits and argument validation are essential to prevent runaway execution.
screen_filing_signals can trigger processes with real-world consequences. An uncontrolled agent might start dozens of builds, send mass notifications, or kick off expensive compute jobs. PolicyLayer enforces rate limits and validates arguments to keep execution within safe bounds.
Execute tools trigger processes. Rate-limit and validate arguments to prevent unintended side effects.
{
"version": "1",
"default": "deny",
"tools": {
"screen_filing_signals": {
"limits": [
{
"counter": "screen_filing_signals_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}See the full MetricDuck — Financial Analysis policy for all 22 tools.
These attack patterns abuse exactly the kind of access screen_filing_signals gives an agent. Each links to the full case and the policy that stops it:
Other execute tools across the catalogue. The same approach applies to each: rate-limit and validate the arguments.
Screen companies by signals across all source types — filings, earnings, transcripts, IR events. This is NOT metric screening (use screen_companies for P/E, ROIC, etc.). Screens by verifiable facts, not LLM-generated scores. Available signals: Filing intelligence (from 10-K/10-Q): - tone_cautious: Management tone is cautious/defensive - tone_shifted: Tone shifted more cautious vs prior - material_weakness: Material weakness in internal controls - accounting_aggressiveness: Accounting aggressiveness flagged (aggressive or conservative) - sbc_high: Stock-based compensation > 15% of revenue - has_new_risks: New risk factors vs prior filing - customer_concentration_high: Customer concentration > 20% or elevated risk - covenant_risk: Covenant tight, waiver obtained, or violation - debt_maturity_near: Significant debt maturing within 12 months - guidance_revised: Guidance raised, lowered, or withdrawn - has_material_charge: Material non-recurring charge or write-down - cash_earnings_divergence: Profitable but negative free cash flow - leverage_spike: Debt-to-equity jumped 50%+ year-over-year - margin_compression: Gross margin declined 3%+ year-over-year - dividend_coverage_weak: Dividend coverage below operating cash flow - sbc_unhedged: Stock comp exceeds buybacks (net dilution) - ocf_deterioration: Operating cash flow declined 40%+ YoY - has_fuel_sensitivity: Fuel cost sensitivity quantified in MD&A - has_commodity_sensitivity: Commodity price sensitivity quantified in MD&A - has_tariff_sensitivity: Tariff/trade policy impact quantified in MD&A - mda_has_scale_claims: ≥3 quantified operational scale claims extracted from MD&A narrative (e.g. renewal rates, member counts, comp sales) - segment_breakdown: Per-segment revenue + ratios + names (always emits when segments present) - customer_concentration: Customer concentration payload (always emits when populated) - geographic_concentration: Geographic concentration payload (always emits when populated) - has_segment_growth_outlier: Any segment grew >20% YoY - has_segment_decline_material: Any segment declined >10% YoY - largest_segment_declining: Top-revenue-share segment is declining YoY - segment_concentration_high: Top segment > 70% of total revenue - risk_landscape_breakdown: Top risks array (severity, is_new, change_vs_prior) + counts - exposure_breakdown: Commodity / tariff exposure (qualitative or quantified magnitude) - has_escalated_risk: ≥1 risk escalated vs prior filing - risk_severity_high_count_above_3: >3 risks with severity=high - guidance_breakdown: Forward guidance accuracy + revision direction payload - guidance_lowered: Guidance lowered (extractor or inferred direction) - guidance_raised: Guidance raised (extractor or inferred direction) - debt_profile_breakdown: Debt + covenant + refinancing_risk + computed near_term_pct - purchase_obligations_breakdown: Take-or-pay schedule with total committed sum - commitments_breakdown: Cloud computing + VIE/SPE exposure - near_term_debt_pct_above_30: >30% of total debt maturing within 12 months - has_purchase_obligations_concentrated: Top counterparty > 50% of purchase obligations - sensitivities_breakdown: Quantified MD&A sensitivities array (variable + sensitivity + direction) - scale_claims_breakdown: MD&A scale claims array (capped at 20 by emission order) - tone_breakdown: Management tone payload (overall + change vs prior) - accounting_breakdown: Accounting quality (material weakness + aggressiveness + sbc%) - sbc_breakdown: Stock-based compensation as % of revenue Earnings releases (8-K Item 2.02 and 6-K Ex 99.1, from earnings press releases): - earnings_revenue_grew: Revenue grew year-over-year - earnings_revenue_declined: Revenue declined year-over-year - earnings_margin_expanded: Operating or gross margin expanded vs prior year - earnings_margin_contracted: Operating or gross margin contracted vs prior year - earnings_guidance_raised_8k: Forward guidance raised in earnings release - earnings_guidance_lowered_8k: Forward guidance lowered in earnings release - earnings_guidance_breakdown: Full forward-guidance array (all metrics, directions, and numeric ranges) from the earnings release - earnings_has_special_items: Non-recurring charges or special items disclosed - earnings_accrual_concerning: Accrual quality weak or concerning (cash vs earnings divergence) - earnings_has_capital_return: Shareholder capital returned (buybacks and/or dividends) Transcript (from earnings call Q&A): - transcript_has_guidance: Specific guidance given on earnings call - transcript_has_product_transitions: Product roadmap activity on call (announced/ramping/live/sunset) - transcript_has_prepared_remarks: Prepared remarks available (true for all transcript sources) - transcript_has_analyst_questions: Analyst Q&A captured with topics + firms - transcript_has_macro_responses: Management addressed macro factors (tariffs, rates, FX, supply chain, etc.) - transcript_has_strategic_priorities: Management articulated ranked strategic priorities on call - transcript_has_competitive_mentions: Competitors/partners named on call with context type - transcript_guidance_raised: ≥1 guidance item raised vs prior quarter (from transcript) - transcript_guidance_lowered: ≥1 guidance item lowered vs prior quarter (from transcript) - transcript_has_scale_claims: ≥3 quantified current-state/historical scale claims on earnings call (e.g. '5x throughput YoY', '1B+ subscribers') - transcript_has_revenue_decompositions: Segment-level revenue decomposed into quantified drivers (volume / price / mix / FX / M&A) on call - transcript_has_kpi_disclosures: ≥2 issuer-disclosed operating KPIs on call (own-defined scorecard metrics: vitality index, NPS, NRR, NIM, combined ratio, RPO, book-to-bill, etc.) - transcript_has_capital_allocation_postures: ≥1 forward capital-allocation posture on call (buyback cadence, leverage target, funding-source rationale, capex policy framing, M&A funding, debt issuance, divestment, authorization runway) - transcript_has_scenario_sensitivities: ≥1 conditional scenario-sensitivity statement on call (trigger event + impact on revenue / EBITDA / margin / EPS — e.g., 'if VantageScore conversion happens, revenue −$270M / EBITDA +$160M / margin +380bps') - transcript_has_forward_commits: ≥1 forward commitment surfaced from Q&A (speaker + analyst + verbatim excerpt + topic tags — calendar-anchored deliverables management committed to under analyst pressure, e.g., 'midyear update on Sadara restructuring') - transcript_has_customer_cohort_metrics: ≥1 customer-cohort metric on call — count + threshold + (optional) Y/Y growth-rate triple, OR top-N deal multi-product attach (SaaS/B2B-skewed; e.g., '630 customers >$5M ACV +22% Y/Y', '17 of top 20 deals included 7+ products') - transcript_has_rpo_disclosure: ≥1 RPO/backlog disclosure on call with structured atom (amount + recognition profile when stated; metric_label, period_end_date, composition_keywords, IR-derived speaker/role) — v1 of forward_commitments family (iter071); bookings + capex_commitment routes ship in v1.1 - transcript_qa_concerns_retained: ≥2 analysts left with concerns retained after Q&A - transcript_qa_forward_committed: ≥2 executive responses with forward-looking commitments on call (count-based; transcript_has_forward_commits exposes the underlying instances) IR events (from company press releases): - ir_acquisition: M&A announced/completed - ir_executive_change: C-suite/board changes - ir_dividend_change: Dividend/buyback announcements - ir_guidance_revision: Guidance raised/lowered/initiated - ir_clinical_regulatory: Clinical trial or regulatory milestone - ir_debt_offering: Debt or equity offering announced - ir_commercial_win: Major commercial contract or win - ir_deliveries: Product deliveries/shipments reported - ir_partnership: Strategic partnerships announced DEF 14A proxy statements (from compensation disclosures): - def14a_peer_group: Compensation peer group disclosed (with company names) - def14a_ceo_pay_ratio: CEO pay ratio disclosed Use Cases: - "Which tech companies have cautious management?" -> signals=["tone_cautious"], sectors=["TECH"] - "Does NVDA have any tone_shifted. It is categorised as a Execute tool in the MetricDuck — Financial Analysis MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MetricDuck — Financial Analysis MCP server in PolicyLayer and add a rule for screen_filing_signals: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches MetricDuck — Financial Analysis. Nothing to install.
screen_filing_signals is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the screen_filing_signals rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for screen_filing_signals. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
screen_filing_signals is provided by the MetricDuck — Financial Analysis MCP server (https://mcp.metricduck.com/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 22 MetricDuck — Financial Analysis tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.