name	deep-code-enrichment
description	Guide for running deep code scanning with LLM-generated symbol descriptions.
compatibility	opencode
metadata	{"audience":"agents","purpose":"code-enrichment"}

Workflow

Run hero scan --code first to ensure the code index is current.
Call hero_code with action: unenriched to get a batch of symbols needing descriptions.
For each symbol, read the source context and write a concise one-line description of what it does.
Call hero_enrich with the descriptions as a JSON array.
Repeat steps 2-4 until no more unenriched symbols are returned.

Description Guidelines

One sentence, starts with a verb (e.g. "Handles incoming MCP tool calls and dispatches to handlers")
Focus on what the symbol DOES, not its signature
Be specific to the codebase, not generic
For types/structs: describe what they represent and their role
For functions: describe the primary behavior and key side effects
Skip trivial getters/setters — describe the business purpose instead

Más de este repositorio

mismo repositorio

agent-reliability

HewlettPackard/morpheus-docs

Behavioral rules that make AI coding agents more reliable — verification, self-correction, and scope discipline.

2026-06-0413

debugging-investigation

HewlettPackard/morpheus-docs

Debugging guidance and structured report template for thorough bug investigation, evidence gathering, and root cause documentation.

2026-06-0413

delivery-audit

HewlettPackard/morpheus-docs

Cold-audit a completed delivery — verify the spec was actually delivered, challenge performative claims, and produce a what-landed report.

2026-06-0413

html-mockup-generation

HewlettPackard/morpheus-docs

Guidelines for generating self-contained HTML mockups that look professional and are easy to iterate on.

2026-06-0413

implementation-principles

HewlettPackard/morpheus-docs

Shared implementation guidance for making minimal, correct, maintainable code changes that fit existing systems.

2026-06-0413

roadmap-review

HewlettPackard/morpheus-docs

The doctrine for `/roadmap-review` — interactive triage of roadmap-shape drift, walked one item at a time, with paste-ready resolution phrasing for the active sizing lens.

2026-06-0413

name	deep-code-enrichment
description	Guide for running deep code scanning with LLM-generated symbol descriptions.
compatibility	opencode
metadata	{"audience":"agents","purpose":"code-enrichment"}

Workflow

Run hero scan --code first to ensure the code index is current.
Call hero_code with action: unenriched to get a batch of symbols needing descriptions.
For each symbol, read the source context and write a concise one-line description of what it does.
Call hero_enrich with the descriptions as a JSON array.
Repeat steps 2-4 until no more unenriched symbols are returned.

Description Guidelines

One sentence, starts with a verb (e.g. "Handles incoming MCP tool calls and dispatches to handlers")
Focus on what the symbol DOES, not its signature
Be specific to the codebase, not generic
For types/structs: describe what they represent and their role
For functions: describe the primary behavior and key side effects
Skip trivial getters/setters — describe the business purpose instead