mcpField
S
sdd-explore

sdd-explore

github.com/fearovex-labsDevelopment54 installs
Repository

About sdd-explore

sdd-explore is a read-only skill for investigating an idea or codebase before an SDD change. It creates a cited exploration artifact that later phases use as a required starting point.

Key features

  1. 01Read-only codebase investigation
  2. 02Cited exploration artifact
  3. 03Required SDD phase gate
  4. 04Project-local skill precedence
  5. 05Optional Engram persistence

Use cases

  1. 01Researching a feature before proposing it
  2. 02Examining code before planning changes
  3. 03Recording findings for later SDD phases

MCP support

The workflow can search Engram through Engram MCP and save an artifact pointer there. If Engram MCP is unavailable, it still writes the exploration document to disk.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/fearovex-labs/agent-config --skill sdd-explore

After installing, type/sdd-explorein Claude Code to confirm it loaded.

Details

From the project README

sdd-explore

Investigates and analyzes an idea or area of the codebase before committing to changes.

Triggers: /sdd-explore <topic>, explore, investigate codebase, analyze before changing, research feature

Purpose

EXPLORE is a mandatory gate of any SDD cycle that proceeds past it. It runs before propose, spec, design, tasks, and apply, and produces a persisted, citation-bearing artifact. No downstream phase may begin until EXPLORE has written its artifact to disk. It creates no code and modifies nothing — it only reads and analyzes.

Skill resolution

When the orchestrator launches this sub-agent, it resolves the skill path using:

1. .claude/skills/sdd-explore/SKILL.md     (project-local — highest priority)
2. ~/.claude/skills/sdd-explore/SKILL.md   (global catalog — fallback)

Project-local skills override the global catalog. See docs/SKILL-RESOLUTION.md for the full algorithm.

Claude execution wiring

These notes bind the tool-neutral logic above to the Claude Code harness. They are the EXECUTION layer that does not port to other tools.

  • Step 0 preload — follow skills/_shared/sdd-phase-common.md Section F (Project Context Load) and Section G (Spec Context Preload). Both are non-blocking. The EXPLORE-local override (mandatory attempt to read ai-context/features/<domain>.md and ai-context/architecture.md, emitting a visible INFO: note on absence) is deliberate and does NOT modify _shared/sdd-phase-common.md Section G.
  • Handoff context preload — check for a pre-seeded proposal in engram with mem_search(query: "sdd/<slug>/proposal"); if present, retrieve it via mem_get_observation(id). Log Handoff context loaded from engram: sdd/<slug>/proposal. Non-blocking.
  • Prior Attempts scan — search engram with mem_search(query: "sdd/archive-report", project: "{project}"), then retrieve each related result via mem_get_observation(id) to check outcome.
  • Artifact persistence (Step 8, disk-primary) — per skills/_shared/sdd-phase-common.md Section C: first Write("docs/sdd/{change-name}/explore.md", <full exploration markdown>), then mem_save a pointer + summary with topic_key: sdd/{change-name}/explore, type: architecture, project: {project}. If Engram MCP is unreachable, still write the disk artifact (the source of truth) and skip the engram pointer only.
  • Orchestrator output — return the neutral Output Contract as the JSON block the orchestrator expects; artifacts reports engram:sdd/{change-name}/explore alongside the disk path, and the manual next step is /sdd-propose.

Comments

Markdown supported