Skip to content
Back to skills

Mk 236

ASecurity

Search Term Observation Classification. Use when you need ambiguous items or items with insufficient data; apply this method to the specified inputs and verify the deliverable

  • 259 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added October 23, 2026
ai-agentsgoperformance

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned October 23, 2026

npx -y skills add backtrue/thinkwithblack-skill --skill mk-236 --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mk 236?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Mk 236
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/backtrue-mk-236/badge)](https://www.skillsdirectory.com/skills/backtrue-mk-236)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: mk-236
description: "Search Term Observation Classification. Use when you need ambiguous items or items with insufficient data; apply this method to the specified inputs and verify the deliverable"
---

# MK-236|Search Term Observation Classification

For a regular chat, upload `packs/mk-236.md`; it includes the shared rules, this method, and required references

## Purpose

This Skill delivers: **Ambiguous items or items with insufficient data**

If the account cannot show all terms, do not call the data complete market demand; the same term can mean different things in different service contexts

## Required Inputs

Provide: search terms, keywords, campaign performance, pages, and demand hypotheses

## Method Steps

1. Classify Search Terms as mine, exclude, observe, or content_signal; this Skill handles only observe
2. When meaning is uncertain, data is insufficient, or evidence does not yet support exclusion/increased investment, retain the original Search Term, current metrics, open questions, and the next evidence needed
3. Do not automatically add an ambiguous Search Term as a negative; explicitly label insufficient evidence as evidence_insufficient / low_confidence / needs_more_data
4. Available Search Terms may be restricted by privacy thresholds, so the observation list represents currently visible data, not the full market

## Calculation and Interpretation

Search term reports may not cover the entire market; observe when evidence is insufficient, and do not choose negative keywords from guesses alone

## Required Output Fields

| Order | Field | How to fill it |
|---:|---|---|
| 1 | Unit of analysis | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |
| 2 | Original clues | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |
| 3 | Category definitions | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |
| 4 | Classification result | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |
| 5 | Supporting / counterevidence | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |
| 6 | Unknowns | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |
| 7 | Downstream use | Original evidence, intermediate method result, or explicitly unknown; do not fabricate |

## Verification Cases

**MK-236-C05|One record meets opposing conditions**

Expected: mark contradictory or pending judgment; do not arbitrarily assign the most common category

## Method Sources

Original file: Google Ads Search AI Skill|Codex 開發交接規格

Source location: observe

Source basis: Specification reference

Original URL: https://docs.google.com/document/d/1jOD2aDqCQiEmH-aIm1ql3xpXZv4Px8gW5cT6XBaoamk/edit

English translation of source excerpt retained in the master table:

> • observe|Keep under observation: insufficient data or uncertain meaning

### Additional Source Locations

Source verification record for the original course material

- Original location: observe

Files in this skill

  • SKILL.md3.1 KB
  • assets/check-context.py14.6 KB
  • assets/context-contract.json10.2 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.