hyp context_en.md

hyp context — Generate Project Manifest

Generate a machine-readable project description (YAML/JSON) so AI can understand your entire project with minimal tokens.

Usage

hyp context                              # YAML to stdout
hyp context -f json                      # JSON format
hyp context -o .hyp/manifest.yaml        # Save to file
hyp context -o manifest.json -f json     # JSON to file

Flags

Flag	Description	Default
-o, --output	Output file path	stdout
-f, --format	Output format: yaml or json	yaml

Output Content

version: "1.0"
framework: HypGo
generated_at: "2026-04-03T14:30:00+08:00"
server:
  addr: ":8080"
  protocol: http2
  tls: true
routes:
  - method: POST
    path: /api/users
    handler_names: [controllers.CreateUser]
    summary: "Create user"
    tags: [users]
    input_type: CreateUserReq
    output_type: UserResp
    responses:
      201: "User created"
database:
  driver: postgres
  has_replicas: true

When to Run

Timing	Reason
After project creation	Generate initial manifest
After adding/modifying Schema routes	Update route info
Before running hyp ai-rules	Include route table in AI config files

AutoSync

Server.Start() automatically calls autosync.SyncSafe() to write the manifest to .hyp/context.yaml. Use hyp context manually for preview or different formats.

Token Efficiency

Scenario	Traditional	HypGo Manifest
AI understands API structure	Read handler source (~5,000 tokens)	Read manifest (~500 tokens)

Security

Manifest does NOT contain: passwords, DSN, tokens, API keys. Only route paths, type names, and server config.

LLM Smart Enrichment (--llm)

v0.8.5+ The --llm flag is new in v0.8.5 and is not available in v0.8.1.

# Specify LLM config file (auto-detects config/llm.yaml or .hyp/llm.yaml)
hyp context --llm config/llm.yaml

# Combined with output options
hyp context --llm config/llm.yaml -o .hyp/manifest.yaml

When enabled, Manifest fields like Summary and Tags are automatically filled in by LLM (Ollama / OpenAI / RAG). See manifest_en.md for details.

Related Commands

• hyp ai-rules — Reads manifest and injects into AI config files
• hyp generate — Generated Schema routes appear in manifest

Detailed Documentation

• manifest.md (package doc)