Name: context-degradation
Rating: 36 (1 reviews)
Author: administrakt0r

context-degradation

Language models exhibit predictable degradation patterns as context length increases. Understanding these patterns is essential for diagnosing failures and designing resilient systems.

Quality

33%

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 ./plugins/AI-Agents-Safe-Coding-Skills/skills/context-degradation/SKILL.md

The canonical home for this skill is context-degradation in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

28%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 skill is a well-organized conceptual reference but is overly verbose, explains material Claude already knows, and offers little executable or procedural guidance. It also lacks genuine bundle files despite hinting at a separate reference, weakening progressive disclosure.

Suggestions

Trim conceptual explanations of known phenomena (attention sinks, poisoning mechanics) to lean, decision-relevant summaries, and move deep empirical detail into a real reference file.

Add concrete, executable mitigations (e.g., specific compaction triggers, masking commands, or code snippets) instead of abstract advice like "use compaction".

Provide a short diagnostic workflow with explicit checkpoints (detect degradation -> locate cause -> apply mitigation -> verify) so users can act on the skill procedurally.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose, devoting long passages to concepts Claude already knows (lost-in-middle, attention sinks, poisoning/distraction/confusion/clash) with padded explanatory prose rather than lean guidance.	2 / 5
Actionability	Guidance is mostly high-level ("Place critical information at the beginning or end", "Apply relevance filtering") with a few illustrative snippets but no executable commands or specific procedures, fitting the minimal-concrete-guidance anchor.	2 / 5
Workflow Clarity	There is a rough "When to Use" list and a Four-Bucket guidance section, but no sequenced multi-step workflow and no validation checkpoints; the skill is conceptual rather than procedural.	2 / 5
Progressive Disclosure	Section headers give it some structure, but ~225 lines of conceptually separable material are inlined and the single referenced "Degradation Patterns Reference" points to no real file (no references/ bundle exists).	3 / 5
	Total	9 / 20 Passed

Description

37%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 states a clear domain and purpose but is vague on concrete actions, omits any "Use when" trigger guidance, and leans on technical jargon rather than natural user phrases. It is distinct enough to avoid total generic overlap but would benefit from explicit triggers and named capabilities.

Suggestions

Add a concrete "Use when..." clause naming natural trigger phrases (e.g., long-context failures, lost-in-middle debugging, agent output degradation).

Replace generic verbs ("diagnosing failures", "designing resilient systems") with specific actions the skill performs (e.g., identifies degradation patterns, recommends placement/compaction mitigations).

Include user-natural keywords and synonyms ("long context", "agent degradation", "lost in the middle") to improve trigger term quality.

Dimension	Reasoning	Score
Specificity	The description names the domain ("predictable degradation patterns as context length increases") but its actions are generic ("diagnosing failures and designing resilient systems") with no concrete capabilities, matching the score-2 anchor of naming the domain with minimal actions.	2 / 5
Completeness	It offers a clear "what" (understanding degradation patterns for diagnosing/designing) but no explicit "when" guidance, and the rubric caps a missing "Use when" clause at 3.	3 / 5
Trigger Term Quality	It relies on technical jargon ("degradation patterns", "context length") with only one or two generic phrases; it lacks the natural wording a user would say when they need this skill and has no "Use when" triggers.	2 / 5
Distinctiveness Conflict Risk	The context-degradation niche is somewhat specific but the description overlaps readily with related context skills (context-fundamentals, context-optimization) referenced in the body, so it could still trigger for the wrong skill.	3 / 5
	Total	10 / 20 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: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills/skills/context-degradation/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.