context-compression

This skill should be used when long-running agent sessions need context compression, structured summarization, compaction, token-per-task optimization, or durable handoff summaries that preserve decisions, files, risks, and next actions.

Quality

75%

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-compression/SKILL.md

Quality

Content

67%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.

The content is well-structured and largely actionable, with real bundle references and clear multi-step workflows. Its main weakness is conciseness: substantial conceptual rationale and repeated justifications add length without proportional value, and validation feedback loops could be made more explicit.

Suggestions

Trim conceptual rationale that re-explains why structure prevents information loss (stated in multiple sections) to improve token efficiency.

Make validation checkpoints explicit in the workflows (e.g. a validate-summary-against-artifacts -> fix -> re-check loop) rather than leaving feedback loops implicit.

Move the detailed six-dimension scoring table and probe taxonomy fully into the evaluation-framework reference, keeping only a brief pointer inline.

Dimension	Reasoning	Score
Conciseness	The body is lengthy (~275 lines) and includes conceptual rationale and overlap with material Claude already knows about summarization tradeoffs; it is mostly efficient but has padded explanatory paragraphs that could be tightened (e.g. repeated justifications of why structure matters).	3 / 5
Actionability	Provides concrete structured-summary templates, a compression-trigger comparison table, probe-type table, and a step-by-step anchored iterative procedure; the inline compression_evaluator.py reference backs the guidance, with only minor gaps in copy-paste-ready executable snippets.	4 / 5
Workflow Clarity	Multi-step workflows are clearly sequenced (three-phase compression workflow, step-by-step anchored iterative summarization, trigger selection) with clear phase outputs, though validation/checkpoint feedback loops are mostly implicit rather than explicit validate->fix->retry steps.	4 / 5
Progressive Disclosure	Good structure with clearly signaled one-level-deep references (./references/evaluation-framework.md and scripts/compression_evaluator.py are real and appropriately cited with 'Read when' triggers); the SKILL.md is organized into overview vs detailed topics, with only minor inlining of evaluation detail that could live in the reference.	4 / 5
	Total	15 / 20 Passed

Description

82%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.

The description is strong: it states concrete capabilities, includes explicit 'Use when' trigger guidance, and covers both what and when. Minor gaps in synonym coverage and in-text distinctiveness keep it just below the top anchor on a couple of dimensions.

Suggestions

Add natural trigger synonyms a user might say, such as 'summarize the conversation' or 'running out of context window', to broaden trigger-term coverage.

Add a short boundary phrase in the description itself (e.g. 'not for general token-efficiency tactics like prefix caching') so the description alone signals distinctiveness from adjacent skills.

Dimension	Reasoning	Score
Specificity	Lists several concrete capabilities ('context compression, structured summarization, compaction, token-per-task optimization, durable handoff summaries that preserve decisions, files, risks, and next actions') but they overlap conceptually and lack distinct actions, sitting just below comprehensive coverage.	4 / 5
Completeness	Explicitly answers both 'what' (the listed compression/summarization capabilities) and 'when' via the 'should be used when long-running agent sessions need...' trigger clause with concrete trigger phrases.	5 / 5
Trigger Term Quality	Good coverage of natural phrases users would say ('context compression', 'summarization', 'compaction', 'handoff summaries'), though some synonyms a user might naturally voice (e.g. 'context window', 'summarize the conversation') are absent.	4 / 5
Distinctiveness Conflict Risk	The niche is fairly clear and the body adds explicit boundary guidance against context-optimization/degradation/memory-systems, but the description text itself overlaps somewhat with adjacent summarization/optimization skills, leaving minor conflict risk.	4 / 5
	Total	17 / 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: muratcankoylan/Agent-Skills-for-Context-Engineering
Path: skills/context-compression/SKILL.md
Commit: 6dbe1a1

Reviewed: 19 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.