Profile Product System

Profile Product System

Trigger

Use when the user asks to understand a product, repo, stack, project context, integration surface, or current engineering maturity before planning work.

When To Use

• Starting a new initiative.
• Joining an unfamiliar codebase.
• Creating baseline product, repo, and tech-stack context.
• Preparing inputs for lifecycle mapping or architecture work.

Inputs Inspected

• User-provided product context.
• README and project docs.
• Repo structure and package manifests.
• Configuration examples and deployment hints.
• Existing generated artifacts under .project/.engineering.

Workflow

1. Run or emulate python "${CLAUDE_PLUGIN_ROOT}/scripts/profile-repo.py" --print to collect factual repo shape.
2. Inspect README, docs, package manifests, deployment config, and existing .project/.engineering/profile artifacts.
3. Separate confirmed facts from product assumptions. Mark unknown users, workflows, integrations, constraints, and ownership explicitly.
4. Write the three profile outputs using the template fields and stable YAML keys.
5. Emit action items for missing critical facts such as unknown deploy target, owner, secrets policy, or primary user.
6. Validate generated Markdown/JSON with python "${CLAUDE_PLUGIN_ROOT}/scripts/validate-artifact.py" <artifact paths>.

Outputs

• .project/.engineering/profile/product-system-profile.yaml
• .project/.engineering/profile/tech-stack-profile.yaml
• .project/.engineering/profile/repo-profile.yaml

Required Front Matter

Generated Markdown companion artifacts must include:

• initiative_id
• skill
• created_at
• status
• confidence
• source_artifacts

YAML/JSON profile files must include equivalent top-level metadata where practical.

Uncertainty Handling

• Use unknown rather than guessing.
• Include an unknowns list with why each item matters.
• Cite files inspected for every non-obvious repo or product claim.

Action Items

Record open questions as checklist items or action_items JSON entries so scripts/emit-action-items.py and scripts/sync-ledger.py can ingest them.

Safety Constraints

• Do not infer secrets from local config values.
• Mark unknowns explicitly instead of guessing.
• Do not edit source files during profiling.

Related Agents

• product-discovery-lead
• solution-architect
• repo-hygiene-maintainer