strategic-compact
建议在逻辑间隔处手动压缩上下文，以在任务阶段中保留上下文，而非任意的自动压缩。
31
25%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./docs/zh-CN/skills/strategic-compact/SKILL.mdQuality
Discovery
0%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is a vague, abstract suggestion written in Chinese that fails to specify concrete actions, natural trigger terms, or explicit usage conditions. It reads more like a tip or guideline than a skill description, making it nearly impossible for Claude to correctly select this skill from a pool of available options.
Suggestions
Rewrite the description to list specific concrete actions the skill performs, e.g., 'Summarizes and compresses conversation context at logical breakpoints such as task phase transitions, preserving key decisions and outputs.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the conversation is getting long, when context window limits are approaching, or when the user mentions context compression, summarization, or memory management.'
Use third-person declarative voice and include distinguishing details that separate this from general summarization or memory skills, e.g., specifying that it targets manual (not automatic) compression at user-defined logical intervals.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description is vague and abstract. It mentions 'compressing context at logical intervals' but does not list concrete actions or specific capabilities. It reads more like a general suggestion than a skill description. | 1 / 3 |
Completeness | The description partially addresses 'what' (manual context compression) but is very weak on specifics. There is no 'Use when...' clause or any explicit trigger guidance for when Claude should select this skill. | 1 / 3 |
Trigger Term Quality | The description is in Chinese and uses abstract terms like '压缩上下文' (compress context) and '逻辑间隔' (logical intervals). These are not natural keywords a user would say when seeking help. There are no common trigger terms or variations. | 1 / 3 |
Distinctiveness Conflict Risk | The description is extremely generic — 'context compression' and 'preserving context across task phases' could overlap with many workflow or memory management skills. There are no distinct triggers to differentiate it. | 1 / 3 |
Total | 4 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a reasonable framework for strategic compaction with a useful decision table and concrete hook configuration. However, it's weakened by the missing core script (suggest-compact.js), some verbosity in explaining concepts Claude already understands, and a tangential token optimization section that dilutes the core message. The skill would benefit from including the actual script and trimming peripheral content.
Suggestions
Include the actual `suggest-compact.js` script content or provide it as a bundle file — without it, the hook setup is not actionable.
Move the 'Token Optimization Patterns' section to a separate reference file (e.g., TOKEN-OPTIMIZATION.md) and link to it, keeping SKILL.md focused on the core compaction workflow.
Add a post-compaction verification step, such as checking that TodoWrite tasks and key file references survived the compaction before proceeding.
Remove or significantly trim the 'Why Strategic Compaction?' section — Claude understands the tradeoffs; a single sentence suffices.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill contains useful information but is more verbose than necessary. The 'Why Strategic Compaction?' section explains concepts Claude can infer, and some tables (like 'What's Retained vs Lost') restate knowledge Claude already has about how compaction works. The token optimization patterns section at the end feels tangential to the core skill. | 2 / 3 |
Actionability | The hook setup JSON is concrete and copy-paste ready, and the decision table provides useful guidance. However, the core script `suggest-compact.js` is referenced but never provided, making the skill incomplete — the user cannot actually implement this without the script. The token optimization tools mentioned (token-optimizer MCP, context-mode) are referenced without concrete setup instructions. | 2 / 3 |
Workflow Clarity | The decision table for when to compact is well-structured and the best practices list provides a clear sequence. However, there's no explicit validation or feedback loop — no way to verify that compaction preserved the right context, no recovery steps if important context was lost, and no checkpoint to confirm readiness before proceeding post-compact. | 2 / 3 |
Progressive Disclosure | The content is organized with clear headers and tables, but it's monolithic — the token optimization patterns section and context composition awareness content could be split into separate reference files. The 'Related' links at the bottom reference external resources but one link appears to be a Twitter/X URL of questionable permanence, and the referenced skills (continuous-learning, memory persistence hooks) aren't provided in the bundle. | 2 / 3 |
Total | 8 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- affaan-m/everything-claude-code
- Commit
928076c
- Reviewed
Table of Contents
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.