dcf · diff

git:20260614.5208b4c to git:20260615.1734111

38 added, 42 removed. Audit A to A.

---

name: dcf

multi_ticker_semantics: target_with_optional_peers

description: DCF valuation model, discounted cash flow, intrinsic value, WACC calculation, terminal value, free cash flow projection, equity value per share, DCF sensitivity analysis, unlevered free cash flow, present value calculation, build a DCF

temporal_scope:

default_quarters: 4

max_quarters: 12

description: "Typical lookback: 4 quarters, max: 12"

allowed_tools:

- search_companies

- search_xbrl_facts

- get_company_financials

- get_company_profile

- search_earnings_calendar

- list_xbrl_concepts

- retrieval_scope: structured_only

+ - search_documents

+ - search_sec_filings

+ - read_source_outline

+ - read_source_deep_outline

+ - read_source_pages

+ - search_keyword_in_source

+ - search_cross_period

+ - batch_search

+ - get_statement_structure

+ - xlsx-read

+ retrieval_scope: unstructured_document_search

min_tool_diversity: 5

---

## Preflight

- !curl -s -o /dev/null -w "%{http_code}" --max-time 2 https://mcp.agentii.ai/mcp/health 2>/dev/null || echo "UNREACHABLE"

+ Run the canonical pre-flight sequence — MCP health probe, ticker resolution, workspace `style.md` override, memory load, and coverage check. See `contracts/preflight.md`.

- **Agent Call Tracing**: The first tool you call will return a `_run_id` in its result. On every subsequent tool call, include HTTP header `X-Agentii-Trace: agent={skill_name}; parent={caller_name}; instance={instance_label}`. The MCP server will inject run_id, depth, and user_id automatically. When spawning parallel sub-agents of the same type, assign each a unique instance label (e.g., equity-research-1, equity-research-2). See `contracts/x-agentii-trace-header.md` for the full contract.

+ Include the `X-Agentii-Trace` header on every tool call per `contracts/x-agentii-trace-header.md`.

## Triggers

- analyze dcf model

- run dcf model analysis

- produce dcf model report

- dcf model breakdown

- dcf model deep dive

- build a dcf model

- assess dcf model

- quantify dcf model

- compare dcf model across peers

- review dcf model for

- generate dcf model on

- dcf model for investment decision

## Defaults

| Parameter | Default | Notes |

|---|---|---|

| lookback_years | 3 | Historical data window |

| include_peers | false | Whether to surface a peer comparison block |

## Methodology

### Retrieval Scope

This skill performs unstructured document search at scale across SEC filings (10-K, 10-Q, 8-K). The three-layer agent-use-ready retrieval protocol (Document Discovery → Page Map → Deep Read) applies to all unstructured document search at scale.

### Retrieval Strategy

- Follow the retrieval strategy decision tree in `retrieval.md`. This skill uses:

- - Branch (a) for structured financial metrics via `search_xbrl_facts` with `list_xbrl_concepts` pre-condition for unfamiliar concepts. **Before querying XBRL facts, optionally call `get_statement_structure/{ticker}?statement_type=income_statement&fiscal_year=<YYYY>` to retrieve the exact line-item hierarchy from `gold.xbrl_presentation` (3.8M rows) — prevents concept-name hallucination and ensures accurate income statement structure for FCF projection .**

- - Branch (b) for multi-period unstructured queries via `search_cross_period`.

- - Branch (c) for single-period document queries via direct `read_source_outline` → `read_source_pages`.

- - Branch (d) for simple lookups via `get_company_profile` / `search_earnings_calendar`.

+ See `contracts/retrieval.md` for the canonical decision tree; skill-specific retrieval detail is in `references/methodology.md`.

### Temporal Scope

Default: 12 fiscal quarters (max 20). Financial modeling: trailing 12 quarters (3 fiscal years) for long-range projection inputs.

### Tool Allowlist

- See frontmatter `allowed_tools` — 12 tools declared for this vertical.

+ See frontmatter `allowed_tools`.

### Protocol

- 1. Pre-retrieval: call `get_company_fiscal_calendar/{ticker}` to resolve fiscal period format.

- 2. Concept discovery: call `list_xbrl_concepts(query=<term>, ticker=<T>)` for unfamiliar XBRL concepts.

- 3. Retrieval: follow the three-layer protocol —

- - Layer 1: `search_documents` / `search_sec_filings` to discover candidate filings.

- - Layer 2: `read_source_outline` to scan page-level metadata.

- - Layer 2.5 (optional): `search_keyword_in_source` to filter large documents.

- - Layer 3: `read_source_pages` to deep-read only selected pages.

- 4. Evidence-pack handoff: produce `evidence-pack.json` + `evidence-digest.md` per the evidence-pack output contract.

- 5. **xlsx-financials output**: invoke `xlsx-financials` as sub-skill to produce formatted `.xlsx` workbook from `get_statement` data for income statement (FCF projection baseline). Output: `{ticker}/{YYYY-MM-DD_HHMM}_statement-income.xlsx` with calculation arc cross-validation .

+ Step-by-step execution detail is in `references/methodology.md`.

## Deliverable Chain

```

[search_xbrl_facts + get_company_financials] → xlsx_build(spec: dcf) → xlsx_recalc → xlsx_audit(hardcoded_count==0) → [.xlsx output] → (optional) pptx_build(executive-summary) → xlsx_convert(pdf)

```

## Validation Gates

1. **projection horizon**: ≥ 5 years (10 years for secular-trends analysis). *If failed*: If < 5 years: refuse delivery, report actual horizon.

2. **terminal growth rate**: < risk-free rate proxy (current 10Y UST). *If failed*: If terminal_g ≥ rf: flag in assumptions section, note conservatism violation.

3. **WACC components**: WACC = (E/V × Ke) + (D/V × Kd × (1-T) with all components cited to source data. *If failed*: If components uncited: refuse delivery, list missing citations.

4. **hardcoded_count**: == 0 for all cells tagged projection|margin|discount_factor|pv|sensitivity per xlsx_audit output. *If failed*: If hardcoded_count > 0: per the hardcode gate, refuse delivery. Bounce back to analytical-subagent ONCE with audit report.

5. **calculation arc cross-validation **: cross-statement balancing verified against `gold.xbrl_calculations` weights — the DCF free-cash-flow projection and income statement structure MUST align with the filer's reported concept hierarchy. Call `get_statement_structure/{ticker}?statement_type=income_statement&include_calculations=true`. Flag discrepancies ≥1% as audit findings. *If failed*: If material discrepancy (≥1%): flag in audit findings, refuse delivery for discrepancies ≥5%. Tool-diversity is also tracked here: distinct MCP tools used MUST be ≥ `min_tool_diversity` (5); below that, flag as depth-insufficient in Coverage Gaps (a quality signal, not a delivery blocker).

## Tool Fallbacks

- |------|-------------|-----------------|---------------------|

- | `read_source_pages` | SQL error / PROXY_ERROR | Use `search_keyword_in_source(document_id, keyword)` if document_id known; otherwise `search_documents` with same query | "source file unavailable; used keyword search instead" |

- | `list_xbrl_concepts` | Timeout / 503 | Use direct `search_xbrl_facts` with standard US-GAAP concepts (Revenues, NetIncomeLoss, EarningsPerShareDiluted, OperatingIncomeLoss, Assets) | "concept discovery skipped due to timeout; using standard US-GAAP concepts" |

- Tool errors are retried ONCE with the fallback action before escalating to the retrieval gaps failure policy. If both Layer 2 and Layer 3 tools are unavailable, enter document access degradation mode (structured data + metadata only, flag output as degraded).

+ Per-tool failure modes and fallback actions are tabulated in `references/tool-fallbacks.md`.

## Output File

- Write the final deliverable to `{ticker}/{{YYYY-MM-DD_HHMM}}_dcf_{{affix}}.md` .

+ Write the final deliverable to `{ticker}/{YYYY-MM-DD_HHMM}_dcf_{affix}.md` .

## Output Structure

- 1. **Executive Summary** — intrinsic value per share, upside/downside vs. current price, key value drivers, WACC used

- 2. **Key Assumptions** — risk-free rate, equity risk premium, beta, cost of equity (Ke), cost of debt (Kd), target capital structure, WACC, terminal growth rate, projection period (≥5 years per Validation Gate 1)

- 3. **Unlevered Free Cash Flow Projection** — EBIT → NOPAT → D&A add-back → Capex → Working Capital Changes → UFCF for each projection year with YoY growth rates

- 4. **Terminal Value** — Gordon Growth Model: TV = UFCF(n+1) / (WACC - g). Terminal growth rate must be < risk-free rate (Validation Gate 2)

- 5. **Enterprise Value** — PV of projected UFCFs + PV of Terminal Value. Mid-year convention applied where appropriate

- 6. **Equity Value Bridge** — Enterprise Value - Net Debt + Cash - Minority Interest → Equity Value

- 7. **Per-Share Value** — Equity Value / Fully Diluted Shares Outstanding → intrinsic value per share

- 8. **Sensitivity Analysis** — 2-way data table: WACC (rows) × Terminal Growth Rate (columns) → per-share value matrix

- 9. **Calculation Arc Cross-Validation ** — income statement structure verified against `gold.xbrl_calculations` weights; FCF drivers aligned with historical margins from XBRL

- 10. **Coverage Gaps & Citations** — data not retrievable + full citation index in `{ticker} {citation_id} page<N>` format

+ The deliverable is a structured markdown report written to the path in `## Output File`. Full section-by-section template (headings, tables, and field definitions) lives in `references/output-structure.md`. Required elements:

+ 1. **Executive Summary** — headline conclusions (≤200 words).

+ 2. **Core analysis sections** — per this skill's methodology and analyst modes.

+ 3. **Data classification** — tag findings `[FACT]` / `[DEDUCTED]` / `[VIEW]` per `contracts/snapshot-synthesis.md`.

+ 4. **Coverage Gaps & Citations** — inline `/v/` citations are PRIMARY (immediately after each fact); the bottom **Citations** section is a non-duplicative roll-up index.

+ 5. **Output frontmatter** — emit the FR-090 structured block per `contracts/output-frontmatter-schema.md`.

+ **Citations & memory**: follow `contracts/citation-and-memory.md` — ≥1 citation per 200 words; every material fact, table row, and metric is immediately followed by its inline clickable `https://agentii.ai/v/{ticker}/{citation_id}/{N}` link; a bottom **Citations** section provides a non-duplicative roll-up index; the closing TUI reply includes a compact **Key Citations** list (headline 5–10 facts) of clickable `/v/` URLs; and append the run to `agentii.md` per `contracts/agentii-md-schema.md`.

+ ## Memory & Snapshot

+ - **Memory load** (pre-flight): load prior workspace context for the ticker before retrieval — see `contracts/memory-load.md`.

+ - **Structured output frontmatter**: emit the FR-090 block (`key_metrics`, `conclusions`, `facts_count`, `deducted_count`, `views_count`, `citation_count`) per `contracts/output-frontmatter-schema.md`.

+ - **Snapshot synthesis**: after writing the deliverable, update the two-tier snapshot and classify findings as `[FACT]`/`[DEDUCTED]`/`[VIEW]` — see `contracts/snapshot-synthesis.md`.

+ - **Session archival**: record the run under `sessions/{YYYY-MM-DD}/` and update `sessions/INDEX.md` per `contracts/session-format.md`.

+ ## Final Summary (TUI)

+ End the closing chat reply with a compact **Key Citations** list (headline 5–10 facts), each a clickable `https://agentii.ai/v/{ticker}/{citation_id}/{N}` link, so the user can cmd+click straight to the exact SEC page. See `contracts/citation-and-memory.md`.

## Error Handling

|---|---|---|---|

| Missing data | Data API returns empty result set | Widen date range and retry once | "No data available for {ticker} in requested window." |

| Sector mismatch | Peer sector != target sector | Filter out mismatched peers | "Removed {n} peer(s) due to sector mismatch." |