SE

select-anchor-filing

Handles financial data extraction from SEC EDGAR filings with strict audit traceability.

Install

mkdir -p .claude/skills/select-anchor-filing && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12907" && unzip -o skill.zip -d .claude/skills/select-anchor-filing && rm skill.zip

Installs to .claude/skills/select-anchor-filing

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

1. The filing is the single source of truth. Every fact traces to `source_accession_number` + verbatim `source_text`. 2. Fail closed on source ambiguity. Fail open on extraction incompleteness. 3. Facts before judgments. Proposals are raw events. Classification is Skill 8, not Sk
280 chars · catalog descriptionno explicit “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Advanced

Key capabilities

  • Search SEC EDGAR for primary filing types
  • Rank filings by chronology quality
  • Select the primary filing for deal analysis
  • Discover supplementary filing candidates
  • Resolve CIK from target name

How it works

This skill reads a seed entry, resolves the CIK, searches all 6 primary SEC filing types, ranks and selects the primary filing, discovers supplementary candidates, and writes the selection rationale to a JSON file.

Inputs & outputs

You give it
A `deal_slug` and a seed row from `Data/reference/deal_url_seeds.csv`
You get back
A `Data/deals/<slug>/source/source_selection.json` file with primary and supplementary searches, and document selection rationale, or a `source/failure_bundle.j

When to use select-anchor-filing

  • Extracting deal data from filings
  • Auditing financial disclosures
  • Retrieving historical SEC data

About this skill

select-anchor-filing

Design Principles

  1. The filing is the single source of truth. Every fact traces to source_accession_number + verbatim source_text.
  2. Fail closed on source ambiguity. Fail open on extraction incompleteness.
  3. Facts before judgments. Proposals are raw events. Classification is Skill 8, not Skill 5.
  4. Alex's collection instructions are the extraction spec. Used for methodology and taxonomy. Never as a factual source.
  5. master.csv is the review artifact, not the estimation artifact.

Overview

Search SEC EDGAR for all 6 primary filing types by CIK, rank filings by chronology quality, and select the primary filing. Also discover supplementary filing candidates. This is the first skill in the deal-agent pipeline. If no usable filing is found, write a failure bundle and stop -- do not proceed to downstream skills.

Input

  • deal_slug (passed by orchestrator)
  • Seed row from Data/reference/deal_url_seeds.csv matching deal_slug

Output

  • Data/deals/<slug>/source/source_selection.json Contains all primary searches, supplementary searches, and document selection rationale.

Tools

ToolPurpose
Shell (curl)All EDGAR access: company search, filings pages, index pages. Always include SEC User-Agent header.
File writeWrite source_selection.json

All SEC access is via curl in a shell command.

Procedure

  1. Read seed entry from Data/reference/deal_url_seeds.csv for the given deal_slug. Extract target_name, filing_url, and any other seed fields.

  2. Resolve CIK. curl the EDGAR company search page by target_name, extract the CIK from the HTML response. See references/edgar-search.md for URL patterns.

  3. Search all 6 primary filing types by CIK regardless of whether a seed URL exists: DEFM14A, PREM14A, SC 14D-9, SC 13E-3, S-4, SC TO-T. For each type, curl the company filings page and parse results. Log every search and its disposition.

  4. Select primary filing. Apply the filing preference ranking (see references/filing-types.md). Pick whichever filing has the most complete "Background of the Merger/Offer" section. If the seed provides a filing_url, treat it as the likely primary, but a higher-ranked filing discovered in step 3 may override it.

  5. For the primary filing: curl its EDGAR index page, parse the document list (filename, description, size), select the main .htm document. Log the selection rationale and why other documents were rejected.

  6. Discover supplementary candidates. All non-primary filings found in step 3 become supplementary candidates. Additionally search for: SC 13D, DEFA14A, 8-K. Log disposition for each.

  7. Write source/source_selection.json with primary_searches, supplementary_searches, and document_selection. See references/edgar-search.md for the schema.

SEC User-Agent (Required)

All curl commands to sec.gov or efts.sec.gov MUST include:

-H "User-Agent: [email protected] deal-extraction-tool/1.0"

SEC blocks requests without a proper User-Agent header.

Gate

At least one filing in source_selection.json has disposition "selected".

Failure Mode

Fail closed. If no filing has a usable background section after searching all 6 types, write source/failure_bundle.json and stop. Do not proceed to Skill 2.

{
  "reason": "no usable filing found across all 6 primary types",
  "last_successful_step": "cik_resolved",
  "candidates_found": []
}

Required Reading

  1. references/edgar-search.md -- CIK resolution, URL patterns, schema
  2. references/filing-types.md -- type descriptions, preference ranking

When not to use it

  • When no usable filing is found after searching all 6 primary types
  • When the user wants to proceed to downstream skills without a primary filing
  • When the user wants to ignore the SEC User-Agent header requirement

Limitations

  • Fails closed on source ambiguity
  • Requires a proper SEC User-Agent header for all `curl` commands
  • If no usable filing is found, it writes a failure bundle and stops

How it compares

This skill automates the complex process of searching, ranking, and selecting SEC filings based on specific criteria and a predefined preference ranking, ensuring traceability and consistency, unlike manual SEC database searches.

Compared to similar skills

select-anchor-filing side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
select-anchor-filing (this skill)05moReviewAdvanced
literature-review5592moReviewAdvanced
openalex-database487moReviewIntermediate
market-research-reports387moReviewAdvanced

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

literature-review

K-Dense-AI

Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting systematic literature reviews, meta-analyses, research synthesis, or comprehensive literature searches across biomedical, scientific, and technical domains. Creates professionally formatted markdown documents and PDFs with verified citations in multiple citation styles (APA, Nature, Vancouver, etc.).

5591,298

openalex-database

davila7

Query and analyze scholarly literature using the OpenAlex database. This skill should be used when searching for academic papers, analyzing research trends, finding works by authors or institutions, tracking citations, discovering open access publications, or conducting bibliometric analysis across 240M+ scholarly works. Use for literature searches, research output analysis, citation analysis, and academic database queries.

48202

market-research-reports

davila7

Generate comprehensive market research reports (50+ pages) in the style of top consulting firms (McKinsey, BCG, Gartner). Features professional LaTeX formatting, extensive visual generation with scientific-schematics and generate-image, deep integration with research-lookup for data gathering, and multi-framework strategic analysis including Porter's Five Forces, PESTLE, SWOT, TAM/SAM/SOM, and BCG Matrix.

38162

scientific-brainstorming

davila7

Research ideation partner. Generate hypotheses, explore interdisciplinary connections, challenge assumptions, develop methodologies, identify research gaps, for creative scientific problem-solving.

37155

exa-search

benjaminjackson

Search the web for content matching a query with AI-powered semantic search. Use for finding relevant web pages, research papers, news articles, code repositories, or any web content by meaning rather than just keywords.

9106

scientific-critical-thinking

davila7

Evaluate research rigor. Assess methodology, experimental design, statistical validity, biases, confounding, evidence quality (GRADE, Cochrane ROB), for critical analysis of scientific claims.

1888

Search skills

Search the agent skills registry