psychological-warfare-siege-defense

Use when defending a besieged city with limited forces against a superior enemy. Combines divine legitimacy, psychological manipulation, feigned weakness, and surprise attack (e.g., Tian Dan's fire cattle) to shatter enemy morale.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

78%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.

The body is a tight, well-structured four-phase workflow with concrete steps and a verification checkpoint, scoring well across the board. The main gaps are a slightly redundant intro line and verification that is a checklist rather than an explicit feedback loop.

Suggestions

Remove or tighten the opening restatement ("A comprehensive approach to defending a besieged city...") since it duplicates the frontmatter description.

Reframe the Verification section as an explicit validate-then-proceed feedback loop (e.g., "If enemy complacency is NOT confirmed, repeat feigned-weakness steps before attacking") to strengthen workflow clarity for this destructive operation.

Add concrete parameters or thresholds to the most abstract directives (e.g., how many fire cattle, what noise sources) to push actionability from concrete-but-strategic toward fully executable.

Dimension	Reasoning	Score
Conciseness	Largely lean with short bulleted phases and little padding, but the opening line ("A comprehensive approach to defending a besieged city...") restates the description and could be trimmed, keeping it just below a 5.	4 / 5
Actionability	Steps are concrete and specific ("Send old people and women to man the walls", "fire cattle with swords on horns", "Attack at night with overwhelming noise"), giving mostly executable guidance, though several directives remain high-level strategic instructions rather than copy-paste-ready artifacts.	4 / 5
Workflow Clarity	A clear four-phase sequence is followed by a dedicated Verification section with explicit preconditions, but the verification reads as a one-way checklist rather than a validate-fix-retry feedback loop, leaving a minor validation gap relative to the top anchor.	4 / 5
Progressive Disclosure	A single, well-organized SKILL.md under 50 lines with clear section headers and no need for external references, satisfying the simple-skill exception for the top score.	5 / 5
	Total	17 / 20 Passed

Description

87%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: it cleanly states both what the skill does and when to use it, in third person, with distinctive niche triggers. The only weakness is that its listed actions are abstract strategic concepts rather than concrete, comprehensive tasks.

Dimension	Reasoning	Score
Specificity	Names the domain and lists several concrete tactics — "divine legitimacy, psychological manipulation, feigned weakness, and surprise attack" — but the actions are somewhat abstract strategic concepts rather than a comprehensive concrete task list, so it stops short of a 5.	4 / 5
Completeness	Explicitly answers both "what" ("Combines divine legitimacy... to shatter enemy morale") and "when" ("Use when defending a besieged city...") with a concrete trigger phrase, matching the top anchor.	5 / 5
Trigger Term Quality	Natural trigger phrasing ("Use when defending a besieged city with limited forces against a superior enemy") with domain-relevant keywords, but it relies on specialized vocabulary and lacks the synonym breadth of a 5.	4 / 5
Distinctiveness Conflict Risk	A very narrow niche (besieged-city defense, Tian Dan's fire cattle) with distinctive triggers and minimal realistic overlap with other skills.	5 / 5
	Total	18 / 20 Passed

Validation

100%

Warnings & errors only

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: baojie/shiji-kb
Path: kg/ontology/ontology-v1/skus/procedural/skill_174/SKILL.md
Commit: 6b836e3

Reviewed: 7 hours ago

Table of Contents

Discovery Implementation Validation

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.