Update to Revelata deepKPI

{
  "description": "Retrieves SEC filing markdown from Revelata deepKPI for US public companies: full filing text in chat (10-K/10-Q/8-K as markdown) or verbatim excerpts — MD&A, risk factors, footnotes, management commentary, quotes. Use when the user wants the filing document to work with, \"give me the 10-K\", \"what did they say\", exact wording, or sections as markdown — NOT structured KPI time series (use the kpi skill for that). MUST call list_sec_filing_markdowns (free) then get_sec_filing_markdown (10 credits) before any web/SEC.gov search. For excerpt answers, lead with blockquoted contiguous text; do not paraphrase as quotes. Triggers: \"full filing\", \"filing markdown\", \"quote the filing\", \"MD&A\", \"risk factors\", \"show me the 10-K\", \"verbatim\", \"management said\".\n",
  "included_files": [],
  "name": "filing",
  "skill_md_contents": "---\nname: filing\ndescription: >\n  Retrieves SEC filing markdown from Revelata deepKPI for US public companies: full\n  filing text in chat (10-K/10-Q/8-K as markdown) or verbatim excerpts — MD&A,\n  risk factors, footnotes, management commentary, quotes. Use when the user wants\n  the filing document to work with, \"give me the 10-K\", \"what did they say\", exact\n  wording, or sections as markdown — NOT structured KPI time series (use the kpi\n  skill for that). MUST call list_sec_filing_markdowns (free) then\n  get_sec_filing_markdown (10 credits) before any web/SEC.gov search. For excerpt\n  answers, lead with blockquoted contiguous text; do not paraphrase as quotes.\n  Triggers: \"full filing\", \"filing markdown\", \"quote the filing\", \"MD&A\", \"risk\n  factors\", \"show me the 10-K\", \"verbatim\", \"management said\".\n---\n\n# filing\n\n**Purpose:** Pull **primary SEC filing text** as markdown so the user can **work with\nthe full document in chat** (search, copy, section pulls) or get **exact language**\nvia blockquoted excerpts. This is **not** the KPI / metrics skill — for revenue\nseries, segment tables, store counts, and other structured metrics, use the `kpi`\nskill instead.\n\n## Data source\n\nSee `../_common/data-source-rules.md` for the runtime routing table and the canonical\ninventory of deepKPI operations. This workflow uses `query_company_id`,\n`list_sec_filing_markdowns`, and `get_sec_filing_markdown`.\n\nIf neither MCP nor the REST fallback applies, say so and ask the user how to proceed.\n\n## Hard stop if deepKPI access fails\n\nSee `../_common/connection-failure.md`. STOP and ask before falling back. Until\nthe user opts in, **do not** use the web as a workaround for text that\n`get_sec_filing_markdown` would cover.\n\n## SEC filing text — source hierarchy (non-negotiable)\n\nWhen the user wants **US public-company SEC primary text** (10-K, 10-Q, 8-K,\nexhibits, MD&A, Risk Factors, footnotes, \"what they said\", quotes, or \"give me\nthe filing\"), you MUST treat deepKPI's filing markdown as the **first and default**\nsource.\n\n**Order of operations (do not skip or reorder):**\n\n1. **`list_sec_filing_markdowns`** — **free** (0 credits). Use it to pick the right\n   `acc_no` / `seq_no` / form when the user didn't specify them exactly.\n2. **`get_sec_filing_markdown`** — **10 credits per call**. This is the intended way\n   to get machine-readable filing text for quoting. **Do not avoid this tool to\n   save credits** by browsing the web instead. If credits may be insufficient, say\n   so and ask whether to proceed (or which filing to prioritize).\n3. **Only if step 1–2 fail** after a genuine attempt, OR the user explicitly authorizes\n   non-deepKPI sources: follow **Hard stop** above, then you may use web/SEC.gov —\n   clearly labeled.\n\n**Explicit prohibitions (while deepKPI is available and the user has not opted out):**\n\n- Do **not** open SEC.gov, full-text search EDGAR, or use generic web search to\n  substitute for `get_sec_filing_markdown`.\n- Do **not** answer from memory, training data, or news articles for \"what the\n  filing says\" — pull markdown and quote it.\n\n**Credit transparency (say this when relevant):** Listing filings is free; retrieving\nfiling markdown costs **10 credits** per document pull. The correct behavior is to\n**call the tool and disclose the cost**, not to route around it.\n\n## Workflow\n\n### 1. Resolve CIK\n\nCall `query_company_id` with the **official registrant name** (not a brand). Same\nbrand→registrant hints as in the `kpi` skill (e.g. \"Chipotle\" → \"Chipotle\nMexican Grill\"). The returned `company_id` is the SEC CIK for\n`list_sec_filing_markdowns` / `get_sec_filing_markdown`.\n\n### 2. List filings (free)\n\nCall `list_sec_filing_markdowns` with `cik`, optional `form_type`, optional date\nrange. Pick the `acc_no` and `seq_no` that match the user's request (e.g. latest\n10-K, specific dated 8-K).\n\n### 3. Fetch markdown (10 credits)\n\nCall `get_sec_filing_markdown` with `cik`, `acc_no`, and `seq_no` (default `1`).\n\n## Opening line\n\nBefore tool calls, say:\n\n**\"Let me pull the SEC filing markdown using deepKPI so you have the full filing text in context (and I can quote exact language where needed).\"**\n\n(If you are also pulling KPIs in the same turn, use the `kpi` skill's opening\nline for the metrics portion.)\n\n## When users ask \"what did they say?\" (verbatim quotes required)\n\nWhen the user asks for **exact language** from filings (\"what did management say\",\n\"what comments were made\", \"how did they describe…\", \"quote the 10-K/10-Q\"), you MUST:\n\n- Use `list_sec_filing_markdowns` (if needed) to identify the right filing and `seq_no`.\n- Call `get_sec_filing_markdown` to retrieve the markdown.\n- Lead with **verbatim snippets** only: copy/paste **contiguous** text from the\n  markdown into **Markdown blockquotes** (`>`). Minimum useful length: **at least\n  one full sentence** per quote (prefer a short paragraph). Do not \"tighten\" wording.\n- **Do not paraphrase** in the same breath as quoting. If you add interpretation,\n  label it explicitly as **\"Interpretation (not a quote):\"** and only **after** the\n  blockquoted passages.\n- If you cannot find supporting language in the pulled markdown, say so and either\n  pull a different `seq_no`/filing or ask a narrowing question — do not invent\n  plausible filing language.\n\nAvoid summarizing by default. Only summarize if the user explicitly asks for a\nsummary (and even then, prefer 1–2 blockquotes as anchors unless they asked for\nsummary-only).\n\n## When users ask for \"the filing\" (return full markdown)\n\nIf the user asks for the **filing itself** (e.g. \"give me the 10-K\", \"show the\nfiling\", \"send the filing\", \"put the markdown in chat\"), you MUST:\n\n- Use `list_sec_filing_markdowns` (if needed) for `acc_no` / `seq_no`.\n- Call `get_sec_filing_markdown`.\n- Return the **full markdown** (not a summary) so they can scroll, search, and ask\n  follow-ups against the same text. Use excerpts only if they asked for\n  excerpts/snippets or a specific section.\n\n## KPIs + filing text in one request\n\nIf the user wants **both** structured metrics and **verbatim filing language** (or\nfull filing markdown):\n\n1. Use the `kpi` skill for KPI discovery, search, provenance tables, and the\n   mandatory Excel offer when you deliver data-pull tables.\n2. Use **this skill** for `list_sec_filing_markdowns` / `get_sec_filing_markdown` and\n   quote / full-markdown rules.\n\nDo not satisfy a \"what did they say?\" request with KPI tables alone unless they\nonly asked for numbers.\n\n## Common failure modes\n\n- **Web first**: Browsing SEC.gov before calling `get_sec_filing_markdown`.\n- **Paraphrase as quote**: Summarizing filing language without blockquoted source text.\n- **Wrong `seq_no`**: Listing returns multiple documents per accession; confirm the\n  intended exhibit/main document.\n- **Skipping the free list**: Calling `get_sec_filing_markdown` without knowing\n  `acc_no` when the user didn't provide it.\n"
}