Context
/forgeproof:run hardcodes effort: high in the skill frontmatter (skills/run/SKILL.md). High effort produces noticeably better plans (e.g. detecting already-implemented requirements instead of re-implementing them), but costs more latency and tokens — a plan-and-stop phase can take a couple of minutes on a small issue.
Request
Let users tune the effort level used for planning, e.g.:
- an argument:
/forgeproof:run 9 --effort low|medium|high, or
- a documented per-install override
Considerations
- Skill frontmatter is static per plugin version; an argument would need the skill to interpret it (effort is a session/skill-level setting, so feasibility of changing it per-invocation needs investigation against current Claude Code capabilities).
- Default should stay
high — plan quality is the product's value; this is an opt-down for quick/simple issues.
- If per-invocation effort turns out not to be supported by the platform, fallback option: ship the guidance in README for users to fork/patch the frontmatter, and revisit when the platform allows it.
Filed from v1.1.0 hands-on testing feedback.
🤖 Generated with Claude Code
Context
/forgeproof:runhardcodeseffort: highin the skill frontmatter (skills/run/SKILL.md). High effort produces noticeably better plans (e.g. detecting already-implemented requirements instead of re-implementing them), but costs more latency and tokens — a plan-and-stop phase can take a couple of minutes on a small issue.Request
Let users tune the effort level used for planning, e.g.:
/forgeproof:run 9 --effort low|medium|high, orConsiderations
high— plan quality is the product's value; this is an opt-down for quick/simple issues.Filed from v1.1.0 hands-on testing feedback.
🤖 Generated with Claude Code