Imported from XSpoonAi/spoon-bot (AGENTS.md). Install upstream with npx skills add XSpoonAi/spoon-bot. Copyright stays with the author.

spoon-bot Agent Instructions

No Case-Specific Routing

Do not add prompt-specific routes, regex classifiers, hidden forced prompts, or one-off workflow branches just to make a replay, demo, or single user prompt pass.
Preserve the user's original prompt. Do not rewrite it into a different task, inject a case-specific plan, or steer it through a hardcoded skill/game/service path.
When a prompt fails, fix the reusable layer underneath: skill metadata, dynamic tool activation, tool contracts, structured request facts, environment handling, provider errors, or loop/tool orchestration.
Regression tests should assert generic behavior and contracts, not exact prompt strings or named demo repositories unless the feature itself owns that fixture.
Temporary diagnostic logic must stay out of committed production paths. If a narrow probe is needed for debugging, remove it before commit.

# after Install: the listing is in your workspace's routing pool - a plan picks it for its slot curl -s -X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'

Install spoon-bot and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

spoon-bot

About

spoon-bot Agent Instructions

No Case-Specific Routing

Use it

Install spoon-bot and it becomes one more candidate the router can pick - when it fits.

Manifest

Reviews