Gan Planner — Development agent for Claude Code
Expand a one-or-two-sentence brief into acceptance criteria, a file list, and a 5-row scoring rubric.
How to install Gan Planner
Installs to ~/.claude/agents/gufranco-claude-engineering-rules-gan-planner.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/gufranco/claude-engineering-rules/HEAD/agents/gan-planner.md -o ~/.claude/agents/gufranco-claude-engineering-rules-gan-planner.md Restart Claude Code, or start a new session, for it to be picked up.
What Gan Planner does
name: gan-planner description: Expand a one-or-two-sentence brief into acceptance criteria, a file list, and a 5-row scoring rubric. First step of the `/gan` skill loop. Runs once per `/gan` invocation. Returns text only; never writes files. tools:
- Read
- Grep
- Glob model: opus color: cyan
You are the planner half of a Generator-Evaluator harness. The orchestrator calls you once per `/gan` invocation with a brief. You return a plan that the generator can execute and the evaluat
Alternatives in Development
- Suede Graph Flo XR Applier — Minimal Suede Graph Flo XR applier that can run only a host-clamped Bash invocation and return its structured 139 ★
- Mathodology Award Judge — Use for one independent blind judge seat scoring an award-level submission against contest rubric criteria 137 ★
- Verify Claim Faithfulness — A verification agent that independently checks, at the sentence level, whether each cited source actually SUPP 134 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Deal Sourcer
Builds and qualifies the acquisition target list. Maps a sector, discovers companies matching the investment t
Fit Scorer
Scores every company on the target list against the buy box into PRIORITY, WATCH or PASS. Builds a written rub
Dev Executor
Explicit-invocation only (Team Dev, Executor) — dispatched by the Manager to mechanically transliterate an alr
Gan Evaluator
GAN Harness — Evaluator agent. Tests the live running application via Playwright, scores against rubric, and p
Scoring
Score a Notion lead row against the user-editable scoring schema. Reads schema tab + enriched lead, computes w
Intake
Parse a Gmail thread into a new Notion lead row. Extracts name, email, company, and raw message body. Called f
Related Skills
Ac Validation
Review Acceptance Criteria for one or more user stories against completeness and quality standards. Checks hap
Multi Build
Claude Code skill that builds one new feature with several AI models at once: a spec with testable acceptance
Gan Trinity
Three AI agents that check each other — Planner, Generator, Evaluator with contract negotiation. Multi-agent d