analyze-candidate-signal

Analyze recruiter conversation transcripts into proposal-only candidate-momentum packets with exact evidence quotes.

1|Updated Aug 3, 2026
One-click install
npx skills add https://github.com/getyak/talent-signal --skill analyze-candidate-signal-getyak
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: analyze-candidate-signal
Source: https://github.com/getyak/talent-signal/tree/main/plugins/talent-signal/skills/analyze-candidate-signal
Command: npx skills add https://github.com/getyak/talent-signal --skill analyze-candidate-signal-getyak

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Recruiters lose track of decision-relevant details buried in candidate conversations, such as deadlines, competing offers, and work-mode preferences. This Skill converts a supplied transcript into a reviewable, evidence-grounded proposal packet without ever confirming state or executing external actions. ## Core Features & Use Cases - Evidence-bound signal extraction: Detects availability, deadlines, competing offers, relocation requirements, and work-mode preferences, each tied to an exact quote, message ID, and speaker. - Proposal-only output contract: Returns one of four dispositions (propose_action, no_action, clarify, block) with at most one recruiter-owned prepare_question proposal and zero execution authority. - Ambiguity preservation: Flags unresolved identity, speaker, timezone, or conditional signals instead of guessing, and blocks candidate-assessment requests outright. - Use Case: A recruiter pastes a candidate chat where the candidate mentions another offer and a Wednesday decision deadline; the Skill returns a proposed decision_deadline state, the exact source quotes, and one suggested clarifying question about the client's remote-work policy. ## Quick Start Ask the assistant to analyze the attached authorized candidate conversation transcript and return a proposal-only momentum packet with evidence quotes and one safe next step.

Frequently Asked Questions about analyze-candidate-signal

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze a recruiter conversation transcript for candidate signals?▼

Format the conversation as JSON with an authorization object, context, and messages array, then pipe it to scripts/analyze_conversation.py via stdin. The analyzer returns a packet with evidence quotes, proposed temporal states, ambiguities, and at most one recruiter-owned action proposal.

What input format does the conversation analyzer require?▼

Input must be one JSON object containing authorization (kind synthetic or user_authorized, plus purpose), context, and a non-empty messages array. Each message needs a unique non-empty id, a speaker label, and exact text.

Can this tool schedule meetings or update an ATS automatically?▼

No. The analyzer has no execution authority and cannot contact anyone, schedule events, or write to Contacts, Calendar, ATS, CRM, or databases. It only proposes a prepare_question artifact that requires an independent human decision.

Why does the analyzer return clarify instead of an action proposal?▼

It returns clarify when candidate or assignment identity is missing, speaker labels are unknown or uncertain, or a relative date cannot be anchored to a timezone. Ambiguity is preserved rather than resolved by guessing.

Does the analyzer support candidate quality or culture-fit scoring?▼

No. Requests for candidate assessment, culture-fit scoring, personality inference, or acceptance probability are blocked outright. The output packet returns disposition block with no state or action proposal.