escape-from-captivity

Use when captured by enemy forces and seeking to escape. Involves feigning death or injury, observing guards and resources, then seizing a sudden opportunity to commandeer a horse and flee to friendly territory.

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.

A well-structured, concise instruction-only skill with a clear sequenced workflow and explicit verification gates. It loses points on actionability and workflow clarity for vague directions ('direction of safety') and the absence of explicit error-recovery feedback loops.

Suggestions

Replace 'direction of safety' / 'friendly territory' with concrete guidance on how to identify a safe direction (e.g., toward known allied positions or away from enemy lines).

Add an explicit feedback loop: state what to do if a verification checkpoint fails (e.g., if the feigned injury is unconvincing, wait and re-observe rather than committing).

Consider trimming the Overview, which restates the frontmatter description, to reduce redundancy.

Dimension	Reasoning	Score
Conciseness	Lean and mostly efficient with no padding of concepts Claude already knows, but the Overview restates the description and 'Expected Outcomes' states fairly obvious results, minor trim opportunities that keep it at 4 rather than 5.	4 / 5
Actionability	Concrete, actionable instructions ('Pretend to be more severely wounded or dead', 'Mount a nearby horse', 'Push off or disable any rider') with specific sub-steps, but a few vague phrases like 'direction of safety' leave minor gaps, so 4 not 5.	4 / 5
Workflow Clarity	A clear 7-step sequence plus a dedicated Verification section with explicit pre-commitment gates ('before committing to action', 'before standing down'); the cap for missing validation does not apply since validation is present, but explicit error-recovery feedback loops are absent, holding it at 4.	4 / 5
Progressive Disclosure	Under 50 lines, single-purpose, no need for external references, and well-organized into clear sections (Overview, Steps, Key Success Factors, Expected Outcomes, Verification), qualifying for 5 under the simple-skill exception.	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 that explicitly states both the trigger condition and the concrete actions, with a distinct niche and low conflict risk. Slightly limited keyword variety keeps trigger quality and specificity at 4 rather than 5.

Suggestions

Add common synonyms such as 'prisoner', 'POW', 'held captive', or 'break out' to broaden natural trigger coverage.

Tighten 'seizing a sudden opportunity' into more concrete phrasing to lift specificity toward comprehensive coverage.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('feigning death or injury, observing guards and resources, then seizing a sudden opportunity to commandeer a horse and flee'), placing it at the 'several specific actions, minor gaps' anchor rather than 5 because 'seizing a sudden opportunity' is somewhat abstract.	4 / 5
Completeness	Explicitly answers both 'when' ('Use when captured by enemy forces and seeking to escape') and 'what' ('Involves feigning death or injury... commandeer a horse and flee') with concrete trigger phrases, matching the 5 anchor.	5 / 5
Trigger Term Quality	Includes natural phrases a user would say ('captured by enemy forces', 'seeking to escape') with good coverage, but missing common synonyms like 'prisoner', 'POW', 'held captive', or 'break out', so it sits at 4 rather than 5.	4 / 5
Distinctiveness Conflict Risk	A clear, narrow niche (escaping enemy captivity) with distinct triggers and minimal overlap risk with other skills, matching the 5 anchor.	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_210/SKILL.md
Commit: 6b836e3

Reviewed: 8 days 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.