skill-library-mcp 1.0.0

package/skills/context-degradation/SKILL.md

@@ -0,0 +1,238 @@
+---
+name: context-degradation
+description: "Recognize patterns of context failure: lost-in-middle, poisoning, distraction, and clash"
+source: "https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/context-degradation"
+risk: safe
+---
+
+## When to Use This Skill
+
+Recognize patterns of context failure: lost-in-middle, poisoning, distraction, and clash
+
+Use this skill when working with recognize patterns of context failure: lost-in-middle, poisoning, distraction, and clash.
+# Context Degradation Patterns
+
+Language models exhibit predictable degradation patterns as context length increases. Understanding these patterns is essential for diagnosing failures and designing resilient systems. Context degradation is not a binary state but a continuum of performance degradation that manifests in several distinct ways.
+
+## When to Activate
+
+Activate this skill when:
+- Agent performance degrades unexpectedly during long conversations
+- Debugging cases where agents produce incorrect or irrelevant outputs
+- Designing systems that must handle large contexts reliably
+- Evaluating context engineering choices for production systems
+- Investigating "lost in middle" phenomena in agent outputs
+- Analyzing context-related failures in agent behavior
+
+## Core Concepts
+
+Context degradation manifests through several distinct patterns. The lost-in-middle phenomenon causes information in the center of context to receive less attention. Context poisoning occurs when errors compound through repeated reference. Context distraction happens when irrelevant information overwhelms relevant content. Context confusion arises when the model cannot determine which context applies. Context clash develops when accumulated information directly conflicts.
+
+These patterns are predictable and can be mitigated through architectural patterns like compaction, masking, partitioning, and isolation.
+
+## Detailed Topics
+
+### The Lost-in-Middle Phenomenon
+
+The most well-documented degradation pattern is the "lost-in-middle" effect, where models demonstrate U-shaped attention curves. Information at the beginning and end of context receives reliable attention, while information buried in the middle suffers from dramatically reduced recall accuracy.
+
+**Empirical Evidence**
+Research demonstrates that relevant information placed in the middle of context experiences 10-40% lower recall accuracy compared to the same information at the beginning or end. This is not a failure of the model but a consequence of attention mechanics and training data distributions.
+
+Models allocate massive attention to the first token (often the BOS token) to stabilize internal states. This creates an "attention sink" that soaks up attention budget. As context grows, the limited budget is stretched thinner, and middle tokens fail to garner sufficient attention weight for reliable retrieval.
+
+**Practical Implications**
+Design context placement with attention patterns in mind. Place critical information at the beginning or end of context. Consider whether information will be queried directly or needs to support reasoning—if the latter, placement matters less but overall signal quality matters more.
+
+For long documents or conversations, use summary structures that surface key information at attention-favored positions. Use explicit section headers and transitions to help models navigate structure.
+
+### Context Poisoning
+
+Context poisoning occurs when hallucinations, errors, or incorrect information enters context and compounds through repeated reference. Once poisoned, context creates feedback loops that reinforce incorrect beliefs.
+
+**How Poisoning Occurs**
+Poisoning typically enters through three pathways. First, tool outputs may contain errors or unexpected formats that models accept as ground truth. Second, retrieved documents may contain incorrect or outdated information that models incorporate into reasoning. Third, model-generated summaries or intermediate outputs may introduce hallucinations that persist in context.
+
+The compounding effect is severe. If an agent's goals section becomes poisoned, it develops strategies that take substantial effort to undo. Each subsequent decision references the poisoned content, reinforcing incorrect assumptions.
+
+**Detection and Recovery**
+Watch for symptoms including degraded output quality on tasks that previously succeeded, tool misalignment where agents call wrong tools or parameters, and hallucinations that persist despite correction attempts. When these symptoms appear, consider context poisoning.
+
+Recovery requires removing or replacing poisoned content. This may involve truncating context to before the poisoning point, explicitly noting the poisoning in context and asking for re-evaluation, or restarting with clean context and preserving only verified information.
+
+### Context Distraction
+
+Context distraction emerges when context grows so long that models over-focus on provided information at the expense of their training knowledge. The model attends to everything in context regardless of relevance, and this creates pressure to use provided information even when internal knowledge is more accurate.
+
+**The Distractor Effect**
+Research shows that even a single irrelevant document in context reduces performance on tasks involving relevant documents. Multiple distractors compound degradation. The effect is not about noise in absolute terms but about attention allocation—irrelevant information competes with relevant information for limited attention budget.
+
+Models do not have a mechanism to "skip" irrelevant context. They must attend to everything provided, and this obligation creates distraction even when the irrelevant information is clearly not useful.
+
+**Mitigation Strategies**
+Mitigate distraction through careful curation of what enters context. Apply relevance filtering before loading retrieved documents. Use namespacing and organization to make irrelevant sections easy to ignore structurally. Consider whether information truly needs to be in context or can be accessed through tool calls instead.
+
+### Context Confusion
+
+Context confusion arises when irrelevant information influences responses in ways that degrade quality. This is related to distraction but distinct—confusion concerns the influence of context on model behavior rather than attention allocation.
+
+If you put something in context, the model has to pay attention to it. The model may incorporate irrelevant information, use inappropriate tool definitions, or apply constraints that came from different contexts. Confusion is especially problematic when context contains multiple task types or when switching between tasks within a single session.
+
+**Signs of Confusion**
+Watch for responses that address the wrong aspect of a query, tool calls that seem appropriate for a different task, or outputs that mix requirements from multiple sources. These indicate confusion about what context applies to the current situation.
+
+**Architectural Solutions**
+Architectural solutions include explicit task segmentation where different tasks get different context windows, clear transitions between task contexts, and state management that isolates context for different objectives.
+
+### Context Clash
+
+Context clash develops when accumulated information directly conflicts, creating contradictory guidance that derails reasoning. This differs from poisoning where one piece of information is incorrect—in clash, multiple correct pieces of information contradict each other.
+
+**Sources of Clash**
+Clash commonly arises from multi-source retrieval where different sources have contradictory information, version conflicts where outdated and current information both appear in context, and perspective conflicts where different viewpoints are valid but incompatible.
+
+**Resolution Approaches**
+Resolution approaches include explicit conflict marking that identifies contradictions and requests clarification, priority rules that establish which source takes precedence, and version filtering that excludes outdated information from context.
+
+### Empirical Benchmarks and Thresholds
+
+Research provides concrete data on degradation patterns that inform design decisions.
+
+**RULER Benchmark Findings**
+The RULER benchmark delivers sobering findings: only 50% of models claiming 32K+ context maintain satisfactory performance at 32K tokens. GPT-5.2 shows the least degradation among current models, while many still drop 30+ points at extended contexts. Near-perfect scores on simple needle-in-haystack tests do not translate to real long-context understanding.
+
+**Model-Specific Degradation Thresholds**
+| Model | Degradation Onset | Severe Degradation | Notes |
+|-------|-------------------|-------------------|-------|
+| GPT-5.2 | ~64K tokens | ~200K tokens | Best overall degradation resistance with thinking mode |
+| Claude Opus 4.5 | ~100K tokens | ~180K tokens | 200K context window, strong attention management |
+| Claude Sonnet 4.5 | ~80K tokens | ~150K tokens | Optimized for agents and coding tasks |
+| Gemini 3 Pro | ~500K tokens | ~800K tokens | 1M context window, native multimodality |
+| Gemini 3 Flash | ~300K tokens | ~600K tokens | 3x speed of Gemini 2.5, 81.2% MMMU-Pro |
+
+**Model-Specific Behavior Patterns**
+Different models exhibit distinct failure modes under context pressure:
+
+- **Claude 4.5 series**: Lowest hallucination rates with calibrated uncertainty. Claude Opus 4.5 achieves 80.9% on SWE-bench Verified. Tends to refuse or ask clarification rather than fabricate.
+- **GPT-5.2**: Two modes available - instant (fast) and thinking (reasoning). Thinking mode reduces hallucination through step-by-step verification but increases latency.
+- **Gemini 3 Pro/Flash**: Native multimodality with 1M context window. Gemini 3 Flash offers 3x speed improvement over previous generation. Strong at multi-modal reasoning across text, code, images, audio, and video.
+
+These patterns inform model selection for different use cases. High-stakes tasks benefit from Claude 4.5's conservative approach or GPT-5.2's thinking mode; speed-critical tasks may use instant modes.
+
+### Counterintuitive Findings
+
+Research reveals several counterintuitive patterns that challenge assumptions about context management.
+
+**Shuffled Haystacks Outperform Coherent Ones**
+Studies found that shuffled (incoherent) haystacks produce better performance than logically coherent ones. This suggests that coherent context may create false associations that confuse retrieval, while incoherent context forces models to rely on exact matching.
+
+**Single Distractors Have Outsized Impact**
+Even a single irrelevant document reduces performance significantly. The effect is not proportional to the amount of noise but follows a step function where the presence of any distractor triggers degradation.
+
+**Needle-Question Similarity Correlation**
+Lower similarity between needle and question pairs shows faster degradation with context length. Tasks requiring inference across dissimilar content are particularly vulnerable.
+
+### When Larger Contexts Hurt
+
+Larger context windows do not uniformly improve performance. In many cases, larger contexts create new problems that outweigh benefits.
+
+**Performance Degradation Curves**
+Models exhibit non-linear degradation with context length. Performance remains stable up to a threshold, then degrades rapidly. The threshold varies by model and task complexity. For many models, meaningful degradation begins around 8,000-16,000 tokens even when context windows support much larger sizes.
+
+**Cost Implications**
+Processing cost grows disproportionately with context length. The cost to process a 400K token context is not double the cost of 200K—it increases exponentially in both time and computing resources. For many applications, this makes large-context processing economically impractical.
+
+**Cognitive Load Metaphor**
+Even with an infinite context, asking a single model to maintain consistent quality across dozens of independent tasks creates a cognitive bottleneck. The model must constantly switch context between items, maintain a comparative framework, and ensure stylistic consistency. This is not a problem that more context solves.
+
+## Practical Guidance
+
+### The Four-Bucket Approach
+
+Four strategies address different aspects of context degradation:
+
+**Write**: Save context outside the window using scratchpads, file systems, or external storage. This keeps active context lean while preserving information access.
+
+**Select**: Pull relevant context into the window through retrieval, filtering, and prioritization. This addresses distraction by excluding irrelevant information.
+
+**Compress**: Reduce tokens while preserving information through summarization, abstraction, and observation masking. This extends effective context capacity.
+
+**Isolate**: Split context across sub-agents or sessions to prevent any single context from growing large enough to degrade. This is the most aggressive strategy but often the most effective.
+
+### Architectural Patterns
+
+Implement these strategies through specific architectural patterns. Use just-in-time context loading to retrieve information only when needed. Use observation masking to replace verbose tool outputs with compact references. Use sub-agent architectures to isolate context for different tasks. Use compaction to summarize growing context before it exceeds limits.
+
+## Examples
+
+**Example 1: Detecting Degradation**
+```yaml
+# Context grows during long conversation
+turn_1: 1000 tokens
+turn_5: 8000 tokens
+turn_10: 25000 tokens
+turn_20: 60000 tokens (degradation begins)
+turn_30: 90000 tokens (significant degradation)
+```
+
+**Example 2: Mitigating Lost-in-Middle**
+```markdown
+# Organize context with critical info at edges
+
+[CURRENT TASK]                      # At start
+- Goal: Generate quarterly report
+- Deadline: End of week
+
+[DETAILED CONTEXT]                  # Middle (less attention)
+- 50 pages of data
+- Multiple analysis sections
+- Supporting evidence
+
+[KEY FINDINGS]                     # At end
+- Revenue up 15%
+- Costs down 8%
+- Growth in Region A
+```
+
+## Guidelines
+
+1. Monitor context length and performance correlation during development
+2. Place critical information at beginning or end of context
+3. Implement compaction triggers before degradation becomes severe
+4. Validate retrieved documents for accuracy before adding to context
+5. Use versioning to prevent outdated information from causing clash
+6. Segment tasks to prevent context confusion across different objectives
+7. Design for graceful degradation rather than assuming perfect conditions
+8. Test with progressively larger contexts to find degradation thresholds
+
+## Integration
+
+This skill builds on context-fundamentals and should be studied after understanding basic context concepts. It connects to:
+
+- context-optimization - Techniques for mitigating degradation
+- multi-agent-patterns - Using isolation to prevent degradation
+- evaluation - Measuring and detecting degradation in production
+
+## References
+
+Internal reference:
+- [Degradation Patterns Reference](./references/patterns.md) - Detailed technical reference
+
+Related skills in this collection:
+- context-fundamentals - Context basics
+- context-optimization - Mitigation techniques
+- evaluation - Detection and measurement
+
+External resources:
+- Research on attention mechanisms and context window limitations
+- Studies on the "lost-in-middle" phenomenon
+- Production engineering guides from AI labs
+
+---
+
+## Skill Metadata
+
+**Created**: 2025-12-20
+**Last Updated**: 2025-12-20
+**Author**: Agent Skills for Context Engineering Contributors
+**Version**: 1.0.0

package/skills/context-driven-development/SKILL.md

@@ -0,0 +1,400 @@
+---
+name: context-driven-development
+description: Use this skill when working with Conductor's context-driven
+  development methodology, managing project context artifacts, or understanding
+  the relationship between product.md, tech-stack.md, and workflow.md files.
+metadata:
+  version: 1.0.0
+---
+
+# Context-Driven Development
+
+Guide for implementing and maintaining context as a managed artifact alongside code, enabling consistent AI interactions and team alignment through structured project documentation.
+
+## Do not use this skill when
+
+- The task is unrelated to context-driven development
+- You need a different domain or tool outside this scope
+
+## Instructions
+
+- Clarify goals, constraints, and required inputs.
+- Apply relevant best practices and validate outcomes.
+- Provide actionable steps and verification.
+- If detailed examples are required, open `resources/implementation-playbook.md`.
+
+## Use this skill when
+
+- Setting up new projects with Conductor
+- Understanding the relationship between context artifacts
+- Maintaining consistency across AI-assisted development sessions
+- Onboarding team members to an existing Conductor project
+- Deciding when to update context documents
+- Managing greenfield vs brownfield project contexts
+
+## Core Philosophy
+
+Context-Driven Development treats project context as a first-class artifact managed alongside code. Instead of relying on ad-hoc prompts or scattered documentation, establish a persistent, structured foundation that informs all AI interactions.
+
+Key principles:
+
+1. **Context precedes code**: Define what you're building and how before implementation
+2. **Living documentation**: Context artifacts evolve with the project
+3. **Single source of truth**: One canonical location for each type of information
+4. **AI alignment**: Consistent context produces consistent AI behavior
+
+## The Workflow
+
+Follow the **Context → Spec & Plan → Implement** workflow:
+
+1. **Context Phase**: Establish or verify project context artifacts exist and are current
+2. **Specification Phase**: Define requirements and acceptance criteria for work units
+3. **Planning Phase**: Break specifications into phased, actionable tasks
+4. **Implementation Phase**: Execute tasks following established workflow patterns
+
+## Artifact Relationships
+
+### product.md - Defines WHAT and WHY
+
+Purpose: Captures product vision, goals, target users, and business context.
+
+Contents:
+
+- Product name and one-line description
+- Problem statement and solution approach
+- Target user personas
+- Core features and capabilities
+- Success metrics and KPIs
+- Product roadmap (high-level)
+
+Update when:
+
+- Product vision or goals change
+- New major features are planned
+- Target audience shifts
+- Business priorities evolve
+
+### product-guidelines.md - Defines HOW to Communicate
+
+Purpose: Establishes brand voice, messaging standards, and communication patterns.
+
+Contents:
+
+- Brand voice and tone guidelines
+- Terminology and glossary
+- Error message conventions
+- User-facing copy standards
+- Documentation style
+
+Update when:
+
+- Brand guidelines change
+- New terminology is introduced
+- Communication patterns need refinement
+
+### tech-stack.md - Defines WITH WHAT
+
+Purpose: Documents technology choices, dependencies, and architectural decisions.
+
+Contents:
+
+- Primary languages and frameworks
+- Key dependencies with versions
+- Infrastructure and deployment targets
+- Development tools and environment
+- Testing frameworks
+- Code quality tools
+
+Update when:
+
+- Adding new dependencies
+- Upgrading major versions
+- Changing infrastructure
+- Adopting new tools or patterns
+
+### workflow.md - Defines HOW to Work
+
+Purpose: Establishes development practices, quality gates, and team workflows.
+
+Contents:
+
+- Development methodology (TDD, etc.)
+- Git workflow and commit conventions
+- Code review requirements
+- Testing requirements and coverage targets
+- Quality assurance gates
+- Deployment procedures
+
+Update when:
+
+- Team practices evolve
+- Quality standards change
+- New workflow patterns are adopted
+
+### tracks.md - Tracks WHAT'S HAPPENING
+
+Purpose: Registry of all work units with status and metadata.
+
+Contents:
+
+- Active tracks with current status
+- Completed tracks with completion dates
+- Track metadata (type, priority, assignee)
+- Links to individual track directories
+
+Update when:
+
+- New tracks are created
+- Track status changes
+- Tracks are completed or archived
+
+## Context Maintenance Principles
+
+### Keep Artifacts Synchronized
+
+Ensure changes in one artifact reflect in related documents:
+
+- New feature in product.md → Update tech-stack.md if new dependencies needed
+- Completed track → Update product.md to reflect new capabilities
+- Workflow change → Update all affected track plans
+
+### Update tech-stack.md When Adding Dependencies
+
+Before adding any new dependency:
+
+1. Check if existing dependencies solve the need
+2. Document the rationale for new dependencies
+3. Add version constraints
+4. Note any configuration requirements
+
+### Update product.md When Features Complete
+
+After completing a feature track:
+
+1. Move feature from "planned" to "implemented" in product.md
+2. Update any affected success metrics
+3. Document any scope changes from original plan
+
+### Verify Context Before Implementation
+
+Before starting any track:
+
+1. Read all context artifacts
+2. Flag any outdated information
+3. Propose updates before proceeding
+4. Confirm context accuracy with stakeholders
+
+## Greenfield vs Brownfield Handling
+
+### Greenfield Projects (New)
+
+For new projects:
+
+1. Run `/conductor:setup` to create all artifacts interactively
+2. Answer questions about product vision, tech preferences, and workflow
+3. Generate initial style guides for chosen languages
+4. Create empty tracks registry
+
+Characteristics:
+
+- Full control over context structure
+- Define standards before code exists
+- Establish patterns early
+
+### Brownfield Projects (Existing)
+
+For existing codebases:
+
+1. Run `/conductor:setup` with existing codebase detection
+2. System analyzes existing code, configs, and documentation
+3. Pre-populate artifacts based on discovered patterns
+4. Review and refine generated context
+
+Characteristics:
+
+- Extract implicit context from existing code
+- Reconcile existing patterns with desired patterns
+- Document technical debt and modernization plans
+- Preserve working patterns while establishing standards
+
+## Benefits
+
+### Team Alignment
+
+- New team members onboard faster with explicit context
+- Consistent terminology and conventions across the team
+- Shared understanding of product goals and technical decisions
+
+### AI Consistency
+
+- AI assistants produce aligned outputs across sessions
+- Reduced need to re-explain context in each interaction
+- Predictable behavior based on documented standards
+
+### Institutional Memory
+
+- Decisions and rationale are preserved
+- Context survives team changes
+- Historical context informs future decisions
+
+### Quality Assurance
+
+- Standards are explicit and verifiable
+- Deviations from context are detectable
+- Quality gates are documented and enforceable
+
+## Directory Structure
+
+```
+conductor/
+├── index.md              # Navigation hub linking all artifacts
+├── product.md            # Product vision and goals
+├── product-guidelines.md # Communication standards
+├── tech-stack.md         # Technology preferences
+├── workflow.md           # Development practices
+├── tracks.md             # Work unit registry
+├── setup_state.json      # Resumable setup state
+├── code_styleguides/     # Language-specific conventions
+│   ├── python.md
+│   ├── typescript.md
+│   └── ...
+└── tracks/
+    └── <track-id>/
+        ├── spec.md
+        ├── plan.md
+        ├── metadata.json
+        └── index.md
+```
+
+## Context Lifecycle
+
+1. **Creation**: Initial setup via `/conductor:setup`
+2. **Validation**: Verify before each track
+3. **Evolution**: Update as project grows
+4. **Synchronization**: Keep artifacts aligned
+5. **Archival**: Document historical decisions
+
+## Context Validation Checklist
+
+Before starting implementation on any track, validate context:
+
+### Product Context
+
+- [ ] product.md reflects current product vision
+- [ ] Target users are accurately described
+- [ ] Feature list is up to date
+- [ ] Success metrics are defined
+
+### Technical Context
+
+- [ ] tech-stack.md lists all current dependencies
+- [ ] Version numbers are accurate
+- [ ] Infrastructure targets are correct
+- [ ] Development tools are documented
+
+### Workflow Context
+
+- [ ] workflow.md describes current practices
+- [ ] Quality gates are defined
+- [ ] Coverage targets are specified
+- [ ] Commit conventions are documented
+
+### Track Context
+
+- [ ] tracks.md shows all active work
+- [ ] No stale or abandoned tracks
+- [ ] Dependencies between tracks are noted
+
+## Common Anti-Patterns
+
+Avoid these context management mistakes:
+
+### Stale Context
+
+Problem: Context documents become outdated and misleading.
+Solution: Update context as part of each track's completion process.
+
+### Context Sprawl
+
+Problem: Information scattered across multiple locations.
+Solution: Use the defined artifact structure; resist creating new document types.
+
+### Implicit Context
+
+Problem: Relying on knowledge not captured in artifacts.
+Solution: If you reference something repeatedly, add it to the appropriate artifact.
+
+### Context Hoarding
+
+Problem: One person maintains context without team input.
+Solution: Review context artifacts in pull requests; make updates collaborative.
+
+### Over-Specification
+
+Problem: Context becomes so detailed it's impossible to maintain.
+Solution: Keep artifacts focused on decisions that affect AI behavior and team alignment.
+
+## Integration with Development Tools
+
+### IDE Integration
+
+Configure your IDE to display context files prominently:
+
+- Pin conductor/product.md for quick reference
+- Add tech-stack.md to project notes
+- Create snippets for common patterns from style guides
+
+### Git Hooks
+
+Consider pre-commit hooks that:
+
+- Warn when dependencies change without tech-stack.md update
+- Remind to update product.md when feature branches merge
+- Validate context artifact syntax
+
+### CI/CD Integration
+
+Include context validation in pipelines:
+
+- Check tech-stack.md matches actual dependencies
+- Verify links in context documents resolve
+- Ensure tracks.md status matches git branch state
+
+## Session Continuity
+
+Conductor supports multi-session development through context persistence:
+
+### Starting a New Session
+
+1. Read index.md to orient yourself
+2. Check tracks.md for active work
+3. Review relevant track's plan.md for current task
+4. Verify context artifacts are current
+
+### Ending a Session
+
+1. Update plan.md with current progress
+2. Note any blockers or decisions made
+3. Commit in-progress work with clear status
+4. Update tracks.md if status changed
+
+### Handling Interruptions
+
+If interrupted mid-task:
+
+1. Mark task as `[~]` with note about stopping point
+2. Commit work-in-progress to feature branch
+3. Document any uncommitted decisions in plan.md
+
+## Best Practices
+
+1. **Read context first**: Always read relevant artifacts before starting work
+2. **Small updates**: Make incremental context changes, not massive rewrites
+3. **Link decisions**: Reference context when making implementation choices
+4. **Version context**: Commit context changes alongside code changes
+5. **Review context**: Include context artifact reviews in code reviews
+6. **Validate regularly**: Run context validation checklist before major work
+7. **Communicate changes**: Notify team when context artifacts change significantly
+8. **Preserve history**: Use git to track context evolution over time
+9. **Question staleness**: If context feels wrong, investigate and update
+10. **Keep it actionable**: Every context item should inform a decision or behavior