Context Fundamentals

Context Engineering Fundamentals

Context is the complete state available to a language model at inference time — system instructions, tool definitions, retrieved documents, message history, and tool outputs. Context engineering is the discipline of curating the smallest high-signal token set that maximizes the likelihood of desired outcomes. Every paragraph below earns its tokens by teaching a non-obvious technique or providing an actionable threshold.

When to Activate

Activate this skill when:

• Designing new agent systems or modifying existing architectures
• Debugging unexpected agent behavior that may relate to context
• Optimizing context usage to reduce token costs or improve performance
• Onboarding new team members to context engineering concepts
• Reviewing context-related design decisions

Core Concepts

Treat context as a finite attention budget, not a storage bin. Every token added competes for the model's attention and depletes a budget that cannot be refilled mid-inference. The engineering problem is maximizing utility per token against three constraints: the hard token limit, the softer effective-capacity ceiling (typically 60-70% of the advertised window), and the U-shaped attention curve that penalizes information placed in the middle of context.

Apply four principles when assembling context:

1. Informativity over exhaustiveness — include only what matters for the current decision; design systems that can retrieve additional information on demand. 2. Position-aware placement — place critical constraints at the beginning and end of context, where recall accuracy runs 85-95%; the middle drops to 76-82% (the "lost-in-the-middle" effect). 3. Progressive disclosure — load skill names and summaries at startup; load full content only when a skill activates for a specific task. 4. Iterative curation — context engineering is not a one-time prompt-writing exercise but an ongoing discipline applied every time content is passed to the model.

Detailed Topics

The Anatomy of Context

System Prompts Organize system prompts into distinct sections using XML tags or Markdown headers (background, instructions, tool guidance, output format). System prompts persist throughout the conversation, so place the most critical constraints at the beginning and end where attention is strongest.

Calibrate instruction altitude to balance two failure modes. Too-low altitude hardcodes brittle logic that breaks when conditions shift. Too-high altitude provides vague guidance that fails to give concrete signals for desired behavior. Aim for heuristic-driven instructions: specific enough to guide behavior, flexible enough to generalize — for example, numbered steps with room for judgment at each step.

Start minimal, then add instructions reactively based on observed failure modes rather than preemptively stuffing edge cases. Curate diverse, canonical few-shot examples that portray expected behavior instead of listing every possible scenario.

Tool Definitions Write tool descriptions that answer three questions: what the tool does, when to use it, and what it returns. Include usage context, parameter defaults, and error cases — agents cannot disambiguate tools that a human engineer cannot disambiguate either.

Keep the tool set minimal. Consolidate overlapping tools because bloated tool sets create ambiguous decision points and consume disproportionate context after JSON serialization (tool schemas typically inflate 2-3x compared to equivalent plain-text descriptions).

Retrieved Documents Maintain lightweight identifiers (file paths, stored queries, web links) and load data into context dynamically using just-in-time retrieval. This mirrors human cognition — maintain an index, not a copy. Strong identifiers (e.g., customer_pricing_rates.json) let agents locate relevant files even without search tools; weak identifiers (e.g., data/file1.json) force unnecessary loads.

When chunking large documents, split at natural semantic boundaries (section headers, paragraph breaks) rather than arbitrary character limits that sever mid-concept.

Message History Message history serves as the agent's scratchpad memory for tracking progress, maintaining task state, and preserving reasoning across turns. For long-running tasks, it can grow to dominate context usage — monitor and apply compaction before it crowds out active instructions.

Cyclically refine history: once a tool has been called deep in the conversation, the raw result rarely needs to remain verbatim. Replace stale tool outputs with compact summaries or references to reduce low-signal bulk.

Tool Outputs Tool outputs typically dominate context — research shows observations can reach 83.9% of total tokens in agent trajectories. Apply observation masking: replace verbose outputs with compact references once the agent has processed the result. Retain only the five most recently accessed file contents; compress or evict older ones.

Context Windows and Attention Mechanics

The Attention Budget For n tokens, the attention mechanism computes n-squared pairwise relationships. As context grows, the model's ability to maintain these relationships degrades — not as a hard cliff but as a performance gradient. Models trained predominantly on shorter sequences have fewer specialized parameters for context-wide dependencies, creating an effective ceiling well below the nominal window size.

Design for this gradient: assume effective capacity is 60-70% of the advertised window. A 200K-token model starts degrading around 120-140K tokens, and complex retrieval accuracy can drop to as low as 15% at extreme lengths.

Position Encoding Limits Position encoding interpolation extends sequence handling beyond training lengths but introduces degradation in positional precision. Expect reduced accuracy for information retrieval and long-range reasoning at extended contexts compared to performance on shorter inputs.

Progressive Disclosure in Practice Implement progressive disclosure at three levels:

1. Skill selection — load only names and descriptions at startup; activate full skill content on demand. 2. Document loading — load summaries first; fetch detail sections only when the task requires them. 3. Tool result retention — keep recent results in full; compress or evict older results.

Keep the boundary crisp: if a skill or document is activated, load it fully rather than partially — partial loads create confusing gaps that degrade reasoning quality.

Context Quality Versus Quantity

Reject the assumption that larger context windows solve memory problems. Processing cost grows disproportionately with context length — not just linear cost scaling, but degraded model performance beyond effective capacity thresholds. Long inputs remain expensive even with prefix caching.

Apply the signal-density test: for each piece of context, ask whether removing it would change the model's output. If not, remove it. Redundant content does not merely waste tokens — it actively dilutes attention from high-signal content.

Practical Guidance

File-System-Based Access

Agents with filesystem access implement progressive disclosure naturally. Store reference materials, documentation, and data externally. Load files only when the current task requires them. Leverage the filesystem's own structure as metadata: file sizes suggest complexity, naming conventions hint at purpose, timestamps serve as proxies for relevance.

Hybrid Context Strategies

Pre-load stable context for speed (CLAUDE.md files, project rules, core instructions) but enable autonomous exploration for dynamic content. The decision boundary depends on content volatility:

• Low volatility (project conventions, team standards): pre-load at session start.
• High volatility (code state, external data, user-specific info): retrieve just-in-time to avoid stale context.

For complex multi-hour tasks, maintain a structured notes file (e.g., NOTES.md) that the agent updates as it works. This enables coherence across context resets without keeping everything in the active window.

Context Budgeting

Allocate explicit budgets per component and monitor during development. Implement compaction triggers at 70-80% utilization — do not wait for the window to fill. Design systems that degrade gracefully: when compaction fires, preserve architectural decisions, unresolved bugs, and implementation details while discarding redundant outputs.

For sub-agent architectures, enforce a compression ratio: a sub-agent may explore using tens of thousands of tokens but must return a condensed summary of 1,000-2,000 tokens. This converts exploration breadth into context-efficient results.

Examples

Example 1: Organizing System Prompts

<BACKGROUND_INFORMATION>
You are a Python expert helping a development team.
Current project: Data processing pipeline in Python 3.9+
</BACKGROUND_INFORMATION>

<INSTRUCTIONS>
- Write clean, idiomatic Python code
- Include type hints for function signatures
- Add docstrings for public functions
- Follow PEP 8 style guidelines
</INSTRUCTIONS>

<TOOL_GUIDANCE>
Use bash for shell operations, python for code tasks.
File operations should use pathlib for cross-platform compatibility.
</TOOL_GUIDANCE>

<OUTPUT_DESCRIPTION>
Provide code blocks with syntax highlighting.
Explain non-obvious decisions in comments.
</OUTPUT_DESCRIPTION>

Example 2: Progressive Document Loading

# Instead of loading all documentation at once:

# Step 1: Load summary
docs/api_summary.md          # Lightweight overview

# Step 2: Load specific section as needed
docs/api/endpoints.md        # Only when API calls needed
docs/api/authentication.md   # Only when auth context needed

Guidelines

1. Treat context as a finite resource with diminishing returns 2. Place critical information at attention-favored positions (beginning and end) 3. Use progressive disclosure to defer loading until needed 4. Organize system prompts with clear section boundaries 5. Monitor context usage during development 6. Implement compaction triggers at 70-80% utilization 7. Design for context degradation rather than hoping to avoid it 8. Prefer smaller high-signal context over larger low-signal context

Gotchas

1. Nominal window is not effective capacity: A model advertising 200K tokens begins degrading around 120-140K. Budget for 60-70% of the nominal window as usable capacity. Exceeding this threshold causes sudden accuracy drops, not gradual degradation — test at realistic context sizes, not toy examples.

2. Character-based token estimates silently drift: The ~4 characters/token heuristic for English prose breaks down for code (2-3 chars/token), URLs and file paths (each slash, dot, and colon is a separate token), and non-English text (often 1-2 chars/token). Use the provider's actual tokenizer (e.g., tiktoken for OpenAI models, Anthropic's token counting API) for any budget-critical calculation.

3. Tool schemas inflate 2-3x after JSON serialization: A tool definition that looks compact in source code expands significantly when serialized — brackets, quotes, colons, and commas each consume tokens. Ten tools with moderate schemas can consume 5,000-8,000 tokens before a single message is sent. Audit serialized tool token counts, not source-code line counts.

4. Message history balloons silently in agentic loops: Each tool call adds both the request and the full response to history. After 20-30 iterations, history can consume 70-80% of the window while the agent shows no visible symptoms until reasoning quality collapses. Set a hard token ceiling on history and trigger compaction proactively.

5. Critical instructions in the middle get lost: The U-shaped attention curve means the middle of context receives 10-40% less recall accuracy than the beginning and end. Never place safety constraints, output format requirements, or behavioral guardrails in the middle of a long system prompt — anchor them at the top or bottom.

6. Progressive disclosure that loads too eagerly defeats its purpose: Loading every "potentially relevant" skill or document at the first hint of relevance recreates the context-stuffing problem. Set strict activation thresholds — a skill should load only when the task explicitly matches its trigger conditions, not when the topic is merely adjacent.

7. Mixing instruction altitudes causes inconsistent behavior: Combining hyper-specific rules ("always use exactly 3 bullet points") with vague directives ("be helpful") in the same prompt creates conflicting signals. Group instructions by altitude level and keep each section internally consistent — either heuristic-driven or prescriptive, not both interleaved.

Integration

This skill provides foundational context that all other skills build upon. It should be studied first before exploring:

• context-degradation - Understanding how context fails
• context-optimization - Techniques for extending context capacity
• multi-agent-patterns - How context isolation enables multi-agent systems
• tool-design - How tool definitions interact with context

References

Internal reference:

• Context Components Reference - Read when: debugging a specific context component (system prompts, tool definitions, message history, tool outputs) or implementing chunking, observation masking, or budget allocation tables

Related skills in this collection:

• context-degradation - Read when: agent performance drops as conversations grow or context fills beyond 60% capacity
• context-optimization - Read when: token costs are too high or compaction/compression strategies are needed

External resources:

• Anthropic's "Effective Context Engineering for AI Agents" — production patterns for compaction, sub-agents, and hybrid retrieval
• Research on transformer attention mechanisms and the lost-in-the-middle effect
• Tokenomics research on agentic software engineering token distribution

---

Skill Metadata

Created: 2025-12-20 Last Updated: 2026-03-17 Author: Agent Skills for Context Engineering Contributors Version: 2.0.0

Context Components: Technical Reference

This document provides detailed technical reference for each context component in agent systems.

System Prompt Engineering

Section Structure

Organize system prompts into distinct sections with clear boundaries. A recommended structure:

<BACKGROUND_INFORMATION>
Context about the domain, user preferences, or project-specific details
</BACKGROUND_INFORMATION>

<INSTRUCTIONS>
Core behavioral guidelines and task instructions
</INSTRUCTIONS>

<TOOL_GUIDANCE>
When and how to use available tools
</TOOL_GUIDANCE>

<OUTPUT_DESCRIPTION>
Expected output format and quality standards
</OUTPUT_DESCRIPTION>

This structure allows agents to locate relevant information quickly and enables selective context loading in advanced implementations.

Altitude Calibration

The "altitude" of instructions refers to the level of abstraction. Consider these examples:

Too Low (Brittle):

If the user asks about pricing, check the pricing table in docs/pricing.md.
If the table shows USD, convert to EUR using the exchange rate in
config/exchange_rates.json. If the user is in the EU, add VAT at the
applicable rate from config/vat_rates.json. Format the response with
the currency symbol, two decimal places, and a note about VAT.

Too High (Vague):

Help users with pricing questions. Be helpful and accurate.

Optimal (Heuristic-Driven):

For pricing inquiries:
1. Retrieve current rates from docs/pricing.md
2. Apply user location adjustments (see config/location_defaults.json)
3. Format with appropriate currency and tax considerations

Prefer exact figures over estimates. When rates are unavailable,
say so explicitly rather than projecting.

The optimal altitude provides clear steps while allowing flexibility in execution.

Tool Definition Specification

Schema Structure

Each tool should define:

{
    "name": "tool_function_name",
    "description": "Clear description of what the tool does and when to use it",
    "parameters": {
        "type": "object",
        "properties": {
            "param_name": {
                "type": "string",
                "description": "What this parameter controls",
                "default": "reasonable_default_value"
            }
        },
        "required": ["param_name"]
    },
    "returns": {
        "type": "object",
        "description": "What the tool returns and its structure"
    }
}

Description Engineering

Tool descriptions should answer: what the tool does, when to use it, and what it produces. Include usage context, examples, and edge cases.

Weak Description:

Search the database for customer information.

Strong Description:

Retrieve customer information by ID or email.

Use when:
- User asks about a specific customer's details, history, or status
- User provides a customer identifier and needs related information

Returns customer object with:
- Basic info (name, email, account status)
- Order history summary
- Support ticket count

Returns null if customer not found. Returns error if database unreachable.

Retrieved Document Management

Identifier Design

Design identifiers that convey meaning and enable efficient retrieval:

Poor identifiers:

• data/file1.json
• ref/ref.md
• 2024/q3/report

Strong identifiers:

• customer_pricing_rates.json
• engineering_onboarding_checklist.md
• 2024_q3_revenue_report.pdf

Strong identifiers allow agents to locate relevant files even without search tools.

Document Chunking Strategy

For large documents, chunk strategically to preserve semantic coherence:

# Pseudocode for semantic chunking
def chunk_document(content):
    """Split document at natural semantic boundaries."""
    boundaries = find_section_headers(content)
    boundaries += find_paragraph_breaks(content)
    boundaries += find_logical_breaks(content)
    
    chunks = []
    for i in range(len(boundaries) - 1):
        chunk = content[boundaries[i]:boundaries[i+1]]
        if len(chunk) > MIN_CHUNK_SIZE and len(chunk) < MAX_CHUNK_SIZE:
            chunks.append(chunk)
    
    return chunks

Avoid arbitrary character limits that split mid-sentence or mid-concept.

Message History Management

Turn Representation

Structure message history to preserve key information:

{
    "role": "user" | "assistant" | "tool",
    "content": "message text",
    "reasoning": "optional chain-of-thought",
    "tool_calls": [list if role="assistant"],
    "tool_output": "output if role="tool"",
    "summary": "compact summary if conversation is long"
}

Summary Injection Pattern

For long conversations, inject summaries at intervals:

def inject_summaries(messages, summary_interval=20):
    """Inject summaries at regular intervals to preserve context."""
    summarized = []
    for i, msg in enumerate(messages):
        summarized.append(msg)
        if i > 0 and i % summary_interval == 0:
            summary = generate_summary(summarized[-summary_interval:])
            summarized.append({
                "role": "system",
                "content": f"Conversation summary: {summary}",
                "is_summary": True
            })
    return summarized

Tool Output Optimization

Response Formats

Provide response format options to control token usage:

def get_customer_response_format():
    return {
        "format": "concise | detailed",
        "fields": ["id", "name", "email", "status", "history_summary"]
    }

The concise format returns essential fields only; detailed returns complete objects.

Observation Masking

For verbose tool outputs, consider masking patterns:

def mask_observation(output, max_length=500):
    """Replace long observations with compact references."""
    if len(output) <= max_length:
        return output
    
    reference_id = store_observation(output)
    return f"[Previous observation elided. Full content stored at reference {reference_id}]"

This preserves information access while reducing token usage.

Context Budget Estimation

Token Counting Approximation

For planning purposes, estimate tokens at approximately 4 characters per token for English text:

1000 words ≈ 7500 characters ≈ 1800-2000 tokens

This is a rough approximation; actual tokenization varies by model and content type.

Context Budget Allocation

Allocate context budget across components:

Monitor actual usage during development to establish baseline allocations.

Progressive Disclosure Implementation

Skill Activation Pattern

def activate_skill_context(skill_name, task_description):
    """Load skill context when task matches skill description."""
    skill_metadata = load_all_skill_metadata()
    
    relevant_skills = []
    for skill in skill_metadata:
        if skill_matches_task(skill, task_description):
            relevant_skills.append(skill)
    
    # Load full content only for most relevant skills
    for skill in relevant_skills[:MAX_CONCURRENT_SKILLS]:
        skill_context = load_skill_content(skill)
        inject_into_context(skill_context)

Reference Loading Pattern

def get_reference(file_reference):
    """Load reference file only when explicitly needed."""
    if not file_reference.is_loaded:
        file_reference.content = read_file(file_reference.path)
        file_reference.is_loaded = True
    return file_reference.content

This pattern ensures files are loaded once and cached for the session.