ericrisco/rsc-harness / skills/fal
fal
VerifiedUse when calling a fal.ai endpoint by id to generate image, audio, or video from JS/Python/curl: subscribe vs submit, queue states, ED25519 webhook signature verification, per-call cost, or migrating off @fal-ai/serverless-client. NOT which model or art direction (that is ai-media); NOT the same models on another platform (that is replicate).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 8.5 KB |
| evals/README.md | markdown | 1.1 KB |
| evals/cases.yaml | config | 2.4 KB |
| references/models-and-cost.md | markdown | 2.7 KB |
| references/queue-and-webhooks.md | markdown | 6.0 KB |
| scripts/verify.shexecutable | script | 3.1 KB |
evals/README.md
View on GitHub# Evals — fal
These cases are run by the repo's eval harness: it loads `cases.yaml` and scores the skill's `description` + body against the `should_trigger` / `should_not_trigger` routing and the `capability` rubric. The `should_trigger` prompts (including the non-obvious "subscribe vs submit" and "signature won't verify" ones, and the Spanish webhook prompt) check that fal fires when an agent is wiring a fal endpoint call. The `should_not_trigger` prompts check the boundaries — each `route_to` id (`ai-media`, `replicate`, `runpod`, `modal`) must exist as a sibling skill for the routing to resolve. Because fal emits a checkable artifact (runnable client code + a webhook handler), the `capability` case is the real rigor check: the produced flow must hit every `must_include` item — current client + `FAL_KEY`, `submit` + `webhook_url` for the long job with the why, the webhook payload shape, ED25519 verification against the JWKS, an idempotent handler, reading the output URL, and a per-second cost estimate with the batch path. Run the harness from the repo root against `skills/fal/`.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |