context-optimization

Context optimization extends the effective capacity of limited context windows through strategic compression, masking, caching, and partitioning. The goal is not to magically increase context windows but to make better use of available capacity.

Quality

36%

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 ./skills/context-optimization/SKILL.md

Quality

Content

38%Scale 1-5

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

This skill reads more like a conceptual tutorial or knowledge base article than an actionable skill for Claude. It spends significant tokens explaining concepts Claude already knows (what KV-cache is, what compaction means) while providing only pseudocode examples that can't be directly executed. The content would benefit greatly from being restructured into a concise overview with detailed techniques split into separate referenced files, and from replacing conceptual explanations with concrete, executable implementations.

Suggestions

Remove explanatory sections like 'What is Compaction', 'The Observation Problem', and 'Understanding KV-Cache'—Claude knows these concepts. Replace with terse actionable guidance.

Provide executable, complete code examples rather than pseudocode with undefined functions like compact_context() and extract_key(). Show actual implementation patterns.

Split detailed technique descriptions (compaction, masking, KV-cache, partitioning) into separate bundle files and keep SKILL.md as a concise overview with clear references.

Add explicit validation/verification steps: e.g., after compaction, verify key information is preserved by checking specific assertions against the summary.

Dimension	Reasoning	Score
Conciseness	Significantly verbose with extensive explanations of concepts Claude already understands (what KV-cache is, what compaction means, what observation masking is). Sections like 'The Observation Problem' and 'Understanding KV-Cache' explain fundamentals rather than providing actionable guidance. The 'Core Concepts' paragraph essentially restates the section headers that follow. Metadata at the bottom wastes tokens.	2 / 5
Actionability	The three code examples are pseudocode-level Python snippets that illustrate concepts but aren't executable in any real system—there are no actual implementations of compact_context(), store_observation(), or extract_key(). The guidance is mostly descriptive ('compaction works by identifying sections that can be compressed') rather than providing concrete, copy-paste-ready implementations or specific tool commands.	3 / 5
Workflow Clarity	The 'Optimization Decision Framework' provides a reasonable decision tree for which technique to apply, and there's a rough sequence (measure → apply compaction → masking → partitioning). However, there are no explicit validation checkpoints or feedback loops for verifying that optimization actually preserved quality. For operations that could degrade context quality (a form of destructive operation), the lack of validation steps caps this at 3.	3 / 5
Progressive Disclosure	The skill is a monolithic ~200-line document with no bundle files to support it. The References section lists internal references and related skills but none are actual files in the bundle. Content like the detailed KV-cache explanation, compaction strategies, and observation masking details could each be separate reference files. Everything is inlined with no actual progressive disclosure structure.	2 / 5
	Total	10 / 20 Passed

Description

33%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.

The description identifies a clear domain (context window optimization) and lists high-level techniques, but lacks concrete actions, natural trigger terms, and an explicit 'when to use' clause. It reads more like a conceptual explanation than a skill selection guide, making it difficult for Claude to reliably choose this skill from a large pool.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases like 'context too long', 'token limit', 'fit more into context', 'reduce prompt size', or 'running out of context'.

Replace or supplement the abstract technique names with concrete actions, e.g., 'Compresses prompts by removing redundant content, partitions long documents into chunks, caches repeated context across turns'.

Include natural synonyms and variations users might use, such as 'token limit', 'prompt too long', 'context length', 'truncation', or 'summarize for context'.

Dimension	Reasoning	Score
Specificity	Names the domain (context optimization) and lists several techniques (compression, masking, caching, partitioning), but these are described at a high level without concrete actions. The second sentence is more of a clarification than additional capability detail.	3 / 5
Completeness	Has a somewhat clear 'what' (extends effective capacity through compression, masking, caching, partitioning) but completely lacks a 'when' clause. There is no explicit trigger guidance or 'Use when...' statement, which per the rubric should cap completeness at 3, and the 'what' is also somewhat vague, bringing it to 2.	2 / 5
Trigger Term Quality	Includes some relevant technical terms like 'context windows', 'compression', 'caching', and 'partitioning', but these are jargon-heavy rather than natural user phrases. Missing natural terms users might say like 'token limit', 'too long', 'reduce context', 'fit more in context', or 'context length'.	2 / 5
Distinctiveness Conflict Risk	The concept of 'context optimization' is somewhat specific, but the techniques listed (compression, caching, partitioning) are broad enough to overlap with other skills related to performance optimization, data processing, or memory management. The domain is identifiable but not sharply delineated.	3 / 5
	Total	10 / 20 Passed

Validation

90%

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 — 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: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: skills/context-optimization/SKILL.md
Commit: 57c135b

Reviewed: 3 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.