Skills context-degradation
📦

context-degradation

Content revision r2 Safe ⚙️ External commands

Diagnose Context Degradation in AI Systems

Long conversations and overloaded prompts can reduce model accuracy in ways that are difficult to diagnose. This skill identifies degradation patterns and recommends practical context management strategies.

Supports: Claude Codex Code(CC)
🥉 78 Bronze

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "context-degradation" from https://skillstore.io/skills/sickn33-context-degradation.md and its manifest at https://skillstore.io/api/skills/sickn33-context-degradation/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/sickn33-context-degradation.md Signed manifest GET /api/skills/sickn33-context-degradation/manifest Signed lockfile GET /api/skills/sickn33-context-degradation/lockfile

Test it

Using "context-degradation". The agent follows an outdated requirement even after the user corrects it.

Expected outcome:

Likely pattern: context poisoning or clash. Remove the outdated requirement, mark the correction as authoritative, and retest from a clean checkpoint.

Using "context-degradation". A model misses a key constraint buried between several long documents.

Expected outcome:

Likely pattern: lost in the middle. Move the constraint near the beginning and restate it in a concise final checklist.

Using "context-degradation". A support agent uses policies from an unrelated product.

Expected outcome:

Likely pattern: context confusion. Separate product contexts, filter retrieved policies by product, and expose only tools relevant to the active task.

Security Audit

Safe
v5 • 7/23/2026 Open versioned report

All four static findings are false positives caused by Markdown examples or ordinary prose. The skill provides conceptual guidance and does not execute commands, access systems, or transmit data.

1
Files scanned
238
Lines analyzed
0
Review items
0
False positives ignored

Risk Factors

⚙️ External commands (3)
SKILL.md:163-170 SKILL.md:170-173 SKILL.md:173-189
No confirmed security findings were detected by the latest completed static and semantic audit. This does not prove the skill has no side effects.
Audited by: codex View Audit History →
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/sickn33-context-degradation/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/sickn33-context-degradation/security.svg)](https://skillstore.io/skills/sickn33-context-degradation?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/sickn33-context-degradation?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/sickn33-context-degradation/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/sickn33-context-degradation.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA · BibTeX · CFF)

APA citation

sickn33. (2026). context-degradation security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/sickn33-context-degradation/audits/5

BibTeX citation

@techreport{sickn33-sickn33-context-degradation-2026, author = {sickn33}, title = {context-degradation security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/sickn33-context-degradation/audits/5}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "context-degradation security audit report (audit version 5)" version: "unspecified" type: report authors: - name: "sickn33" date-released: "2026-07-23" url: "https://skillstore.io/skills/sickn33-context-degradation/audits/5" identifiers: - type: other value: "skillstore:sickn33-context-degradation:audit:5" description: "Skillstore immutable audit report identifier"

Compare variants

4 installable variants

Each author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.

Why this variant is first

Highest Skillstore Score
ChakshuGautam Recommended

chakshugautam-context-degradation

Skillstore Score 81
Evidence Confidence High
Skillstore usage 15
Updated

2026-08-21

muratcankoylan

muratcankoylan-context-degradation

Skillstore Score 80
Evidence Confidence High
Skillstore usage 21
Updated

2026-08-21

Asmayaseen

asmayaseen-context-degradation

Skillstore Score 80
Evidence Confidence High
Skillstore usage 15
Updated

2026-08-21

sickn33 Current

sickn33-context-degradation

Skillstore Score 78
Evidence Confidence High
Skillstore usage 16
Updated

2026-08-21

Skillstore Score

Why this score Evidence Confidence: High
55
Architecture
85
Maintainability
87
Content
69
Community
83
Spec Compliance

What You Can Build

Debug a Long Agent Session

Identify why an agent forgets requirements, repeats errors, or selects irrelevant tools after many turns.

Design Context Evaluations

Create tests that vary context length, information placement, distractors, and conflicting sources.

Improve Prompt Architecture

Choose filtering, compaction, retrieval, or task isolation based on the observed failure pattern.

Try These Prompts

Identify the Failure Pattern 
Review this agent behavior: [behavior]. Classify the likely context degradation pattern and explain the evidence.
Improve Context Placement 
Review this prompt structure: [structure]. Reorder critical information to reduce lost-in-middle risk while preserving all requirements.
Choose a Mitigation Strategy 
Analyze this workflow: [workflow]. Compare write, select, compress, and isolate strategies, then recommend an implementation with tradeoffs.
Design a Degradation Evaluation 
Create an evaluation plan for [system]. Test context length, placement, distractors, poisoning, and conflicting sources. Define metrics, thresholds, and recovery checks.

Best Practices

  • Measure task quality as context grows instead of relying on advertised window size.
  • Keep verified goals and critical constraints near attention-favored positions.
  • Filter, version, and label retrieved information before adding it to context.

Avoid

  • Do not load every available document without relevance filtering.
  • Do not preserve incorrect summaries after evidence changes.
  • Do not mix unrelated tasks, tool definitions, and policies in one context.

Frequently Asked Questions

What is context degradation?
It is a gradual decline in model reliability as context becomes longer, noisier, contradictory, or incorrectly summarized.
What does lost in the middle mean?
Models often recall information near the beginning and end better than information placed in the middle.
How is context poisoning different from context clash?
Poisoning propagates incorrect information. Clash occurs when multiple context items conflict, even when each item may be valid in another setting.
Does a larger context window prevent degradation?
No. Capacity does not guarantee consistent retrieval, reasoning, or instruction adherence across the entire window.
Which mitigation should I try first?
Start by removing irrelevant content and surfacing current goals. Add compaction or task isolation when the context continues growing.
Can this skill test my system automatically?
No. It provides diagnostic and evaluation guidance, but you must run measurements with your model, prompts, tools, and data.

Developer Details

Author

sickn33

License

MIT

Skillstore revision

r2

Version notice

The author did not declare a version.

Repository

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/context-degradation

Ref

f9e2c34b4f19c7f3e6b0a1e93227b5f77cc12526

Maintenance freshness

7/26/2026

Usage

13 downloads · 148 views

File structure

📄 SKILL.md