strategic-compact
Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.
57
48%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/strategic-compact/SKILL.mdQuality
Discovery
32%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 communicates a reasonably specific concept—manual context compaction at logical intervals—but lacks an explicit 'Use when...' clause, which is critical for skill selection. It also misses natural trigger terms users might employ (e.g., 'running out of context', 'long conversation', 'token limit') and doesn't list multiple concrete actions beyond the single suggestion behavior.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the conversation is getting long, the user mentions running out of context, or when transitioning between major task phases.'
Include natural trigger terms users would say, such as 'context window', 'token limit', 'long conversation', 'memory', or 'running out of space'.
List additional concrete actions beyond suggesting compaction, e.g., 'Identifies logical breakpoints in tasks, summarizes completed phases, and recommends when to compact context to preserve key information.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | It names the domain (context compaction) and describes the core action (suggests manual context compaction at logical intervals), but doesn't list multiple concrete actions or elaborate on what specific steps are involved. | 2 / 3 |
Completeness | It describes what the skill does (suggests manual compaction at logical intervals) but lacks any explicit 'Use when...' clause or trigger guidance, which per the rubric should cap completeness at 2, and the 'when' is entirely missing, warranting a 1. | 1 / 3 |
Trigger Term Quality | Includes relevant terms like 'context compaction' and 'auto-compaction', but misses common user-facing variations such as 'context window', 'running out of context', 'memory management', 'long conversation', or 'token limit'. | 2 / 3 |
Distinctiveness Conflict Risk | The concept of 'context compaction' is fairly niche, but the description could overlap with general conversation management or memory-related skills. It's somewhat specific but not sharply delineated. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides genuinely useful guidance on strategic compaction with strong actionability — the hook configuration is copy-paste ready and the decision guide table is well-designed. However, it suffers from scope creep in the 'Token Optimization Patterns' section which adds ~30% more content on tangentially related topics, and the workflow could benefit from explicit validation steps (e.g., confirming important context is saved before compacting).
Suggestions
Remove or extract the 'Token Optimization Patterns' section to a separate file — it covers tangential topics (lazy loading, duplicate detection, external tools) that dilute the core skill's focus on strategic compaction timing.
Add an explicit pre-compaction validation step to the workflow, such as: 'Before compacting, verify critical context is saved to files/memory/TodoWrite — check that your plan, key file paths, and current state are persisted.'
Trim the 'What Survives Compaction' and 'Why Strategic Compaction?' sections — Claude understands these concepts and they can be reduced to 2-3 bullet points each.
| 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, the 'What Survives Compaction' table is largely common knowledge for Claude, and the 'Token Optimization Patterns' section drifts into tangential topics (trigger-table lazy loading, duplicate instruction detection) that aren't core to the strategic compact skill. | 2 / 3 |
Actionability | Provides concrete, copy-paste ready JSON configuration for hooks, a clear decision guide table for when to compact, specific commands like `/compact Focus on implementing auth middleware next`, and actionable best practices. The hook setup is fully executable. | 3 / 3 |
Workflow Clarity | The decision guide table is excellent for deciding when to compact, and the 'How It Works' section explains the hook mechanism. However, there's no explicit end-to-end workflow with validation steps — for example, no guidance on verifying that important context was saved before compacting, or what to do if compaction loses critical state. | 2 / 3 |
Progressive Disclosure | The content is reasonably well-structured with clear headers and tables, but the 'Token Optimization Patterns' section is a large inline block that covers tangential topics (lazy loading, context composition, duplicate detection, external tools) that should either be in a separate reference file or removed. The 'Related' section at the end hints at external references but they're URLs and vague skill names rather than clear file links. | 2 / 3 |
Total | 9 / 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
- ysyecust/everything-claude-code
- Commit
79cc4e3
- 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.