strategic-compact

建议在逻辑间隔进行手动上下文压缩（Context Compaction），以便在任务阶段中保留上下文，而不是依赖任意的自动压缩。

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/strategic-compact/SKILL.md

Quality

Content

65%Weight 40%Scale 1-3

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

The body is actionable and well-structured with concrete hook config and a useful decision table. It is held back by explanatory padding in the rationale section, a missing validation checkpoint around compaction, and a referenced script file that is absent from the bundle.

Suggestions

Tighten the '为什么要进行策略性压缩？' section to remove general explanation of auto-compaction that Claude already knows.

Add an explicit pre-compact validation step (e.g. write key state to TodoWrite/memory, confirm the /compact summary captures it) before invoking /compact.

Either bundle the referenced 'suggest-compact.js' script or add a one-level-deep reference file (e.g. HOOK_REFERENCE.md) so the referenced file actually exists.

Dimension	Reasoning	Score
Conciseness	Content is mostly efficient and well-organized, but the '为什么要进行策略性压缩？' section explains general auto-compaction behavior Claude already knows, adding some unnecessary context.	2 / 3
Actionability	Provides concrete, executable guidance: a ready-to-paste hook configuration JSON, environment variables with defaults, and a decision table with explicit '/compact' usage examples.	3 / 3
Workflow Clarity	The decision table and best-practices list sequence the process, but there is no validation/checkpoint feedback loop (e.g. verify summary retained key state before compacting) for a destructive-ish context-dropping operation.	2 / 3
Progressive Disclosure	The body is reasonably sectioned, but the referenced 'suggest-compact.js' script does not exist as a bundle file and there are no one-level-deep reference links to detailed material, so structure is only partially realized.	2 / 3
	Total	9 / 12 Passed

Description

57%Weight 40%Scale 1-3

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

The description conveys a specific, distinct niche (strategic manual context compaction) but reads as a rationale rather than a trigger. It lacks an explicit 'Use when...' clause and natural user-facing keywords, which cap completeness and trigger-term quality at 2.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when running long multi-stage sessions, switching tasks, or nearing the context limit.'

Include natural trigger terms users would say, such as '/compact', 'long session', 'context limit', 'context window', 'token budget'.

Mention third-person concrete actions ('Suggests strategic /compact points…') instead of framing only as advice.

Dimension	Reasoning	Score
Specificity	It names a concrete action ('建议…手动上下文压缩', Context Compaction) and the domain (context preservation), but does not enumerate multiple specific actions like the score-3 anchor requires.	2 / 3
Completeness	It states what the skill does (recommends manual context compaction at logical intervals) but lacks an explicit 'Use when...' trigger clause telling Claude when to invoke it, capping completeness at 2 per the guidelines.	2 / 3
Trigger Term Quality	Terms like '上下文压缩', 'Context Compaction', '逻辑间隔' are relevant but the phrasing is somewhat technical; common user-facing variations like '/compact', '长会话', '上下文窗口' are not present in the description.	2 / 3
Distinctiveness Conflict Risk	The niche is clearly specific — strategic manual context compaction vs auto-compaction — with distinct triggers that are unlikely to conflict with other skills.	3 / 3
	Total	9 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: xu-xiang/everything-claude-code-zh
Path: .agents/skills/strategic-compact/SKILL.md
Commit: dfbf946

Reviewed: 28 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.