Files
wushu/.trae/skills/planning-with-files/.kiro/skills/planning-with-files/references/manus-principles.md
2026-03-30 02:35:31 +08:00

3.6 KiB

Manus context engineering (reference)

This workflow is inspired by Manus-style context engineering: treat markdown on disk as durable working memory while the model context window behaves like volatile RAM.

Kiro layout: planning files live under .kiro/plan/ (not the project root). See Kiro Steering — file references for #[[file:path]] live includes.


Reference: Manus Context Engineering Principles

This skill is based on context engineering principles from Manus, the AI agent company acquired by Meta for $2 billion in December 2025.

The 6 Manus Principles

Principle 1: Design Around KV-Cache

"KV-cache hit rate is THE single most important metric for production AI agents."

Statistics:

  • ~100:1 input-to-output token ratio
  • Cached tokens: $0.30/MTok vs Uncached: $3/MTok
  • 10x cost difference!

Implementation:

  • Keep prompt prefixes STABLE (single-token change invalidates cache)
  • NO timestamps in system prompts
  • Make context APPEND-ONLY with deterministic serialization

Principle 2: Mask, Don't Remove

Don't dynamically remove tools (breaks KV-cache). Use logit masking instead.

Best Practice: Use consistent action prefixes (e.g., browser_, shell_, file_) for easier masking.

Principle 3: Filesystem as External Memory

"Markdown is my 'working memory' on disk."

The Formula:

Context Window = RAM (volatile, limited)
Filesystem = Disk (persistent, unlimited)

Compression Must Be Restorable:

  • Keep URLs even if web content is dropped
  • Keep file paths when dropping document contents
  • Never lose the pointer to full data

Principle 4: Manipulate Attention Through Recitation

"Creates and updates todo.md throughout tasks to push global plan into model's recent attention span."

Problem: After ~50 tool calls, models forget original goals ("lost in the middle" effect).

Solution: Re-read .kiro/plan/task_plan.md before major decisions. Goals appear in the attention window.

Principle 5: Keep the Wrong Stuff In

"Leave the wrong turns in the context."

Why:

  • Failed actions with stack traces let model implicitly update beliefs
  • Reduces mistake repetition
  • Error recovery is "one of the clearest signals of TRUE agentic behavior"

Principle 6: Don't Get Few-Shotted

"Uniformity breeds fragility."

Problem: Repetitive action-observation pairs cause drift and hallucination.

Solution: Introduce controlled variation:

  • Vary phrasings slightly
  • Don't copy-paste patterns blindly
  • Recalibrate on repetitive tasks

The 3 Context Engineering Strategies

Based on Lance Martin's analysis of Manus architecture.

Strategy 1: Context Reduction

Compaction:

Tool calls have TWO representations:
├── FULL: Raw tool content (stored in filesystem)
└── COMPACT: Reference/file path only

RULES:
- Apply compaction to STALE (older) tool results
- Keep RECENT results FULL (to guide next decision)

Strategy 2: Context Isolation (Multi-Agent)

Multi-agent setups can isolate exploration in separate contexts while persisting shared state in files (e.g. under .kiro/plan/).

Strategy 3: Context Offloading

  • Store full results in the filesystem, not only in context
  • Progressive disclosure: load information only as needed

File Types (Kiro paths)

File Purpose
.kiro/plan/task_plan.md Phase tracking, progress
.kiro/plan/findings.md Discoveries, decisions
.kiro/plan/progress.md Session log

Source

Manus context engineering blog: https://manus.im/blog/Context-Engineering-for-AI-Agents-Lessons-from-Building-Manus