CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

kyc-doc-parse

Parse an investor or client onboarding packet into structured KYC fields — identity, ownership, control, source of funds, and document inventory. Use as the first step of KYC screening; output feeds the rules engine.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is kyc-doc-parse in anthropics/financial-services

SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A lean, well-structured instruction skill: concrete output schema, explicit untrusted-input handling, and a clear three-step sequence with a pre-handoff gap check. The main improvements are a worked example record and a disposition rule for flagged gaps.

Suggestions

Include one filled example JSON record (with realistic values) alongside the template so the expected extraction shape is unambiguous for common cases.

Add a feedback loop for Step 3: state what to do when gaps are flagged (e.g., proceed with nulls and list gaps in a "gaps" field for the rules engine, or request documents before proceeding).

Specify where the Step 3 gap flags live in the output (e.g., a "gaps" array in the JSON record) so the handoff to kyc-rules is concrete.

DimensionReasoningScore

Conciseness

At 49 lines there is no filler: the document-type table and JSON template carry all information with no explanations of concepts Claude already knows, and directives like "Use `null` for any field not found — do not guess" add exactly the non-obvious guidance that earns its tokens.

5 / 5

Actionability

The Step 2 JSON record is a copy-paste-ready output template with concrete field names and types, and Step 1's table enumerates exact document types. The minor gap is that all example values are "..." placeholders — a filled example record would show a realistic extraction.

4 / 5

Workflow Clarity

Three clearly sequenced steps (inventory → extract → flag gaps) with an explicit pre-handoff checkpoint in Step 3 (expired IDs, stale address proof, missing UBO chart). It falls short of a 5 because there is no feedback loop telling the agent what to do when a gap is flagged (proceed, request documents, or halt).

4 / 5

Progressive Disclosure

The skill is under 50 lines, single-purpose, and needs no external references; its sections (untrusted-input note, inventory table, output schema, gap flags) are well-organized, which per the rubric's simple-skill guidance merits a 5.

5 / 5

Total

18

/

20

Passed

Description

83%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: specific, third-person, appropriately concise, and clearly positioned in a KYC pipeline. Its main weakness is trigger breadth — it covers "KYC" and "onboarding" but omits the AML/CDD/UBO vocabulary users in this domain commonly say.

Suggestions

Add natural trigger synonyms users would say, e.g. "Use when the user mentions KYC, AML screening, customer due diligence (CDD), or beneficial ownership (UBO) checks."

Make the "when" clause user-trigger shaped rather than only pipeline-shaped — state the situations that call for this skill (e.g., start of onboarding a new investor/client), not just that it precedes the rules engine.

DimensionReasoningScore

Specificity

"Parse an investor or client onboarding packet into structured KYC fields — identity, ownership, control, source of funds, and document inventory" names the domain and enumerates five concrete field categories, giving comprehensive coverage of the extraction scope.

5 / 5

Completeness

A clear "what" (parse packets into structured KYC fields) is paired with an explicit usage trigger ("Use as the first step of KYC screening"), but the "when" is framed as pipeline position rather than concrete user-side trigger phrases, so it could be more explicit.

4 / 5

Trigger Term Quality

Natural terms like "KYC", "KYC screening", "client onboarding", and "investor" are present, but common synonyms users would say ("AML", "customer due diligence", "CDD", "UBO", "beneficial ownership") are missing — good coverage with a few natural terms absent.

4 / 5

Distinctiveness Conflict Risk

"Structured KYC fields", "onboarding packet", and "output feeds the rules engine" carve a clear niche with distinct triggers; overlap risk with generic document-parsing skills is minimal.

5 / 5

Total

18

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

anthropics/financial-services

Path

plugins/vertical-plugins/operations/skills/kyc-doc-parse/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.