hassan-mohiddin/freeflow / .deprecated/tool-execution-v2/capabilities/tool-execution
tool-execution
VerifiedUse for Freeflow programs, mutations, exact artifact recovery, and unfamiliar catalog operations; bounded direct read/search needs no skill solely for one call.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.5 KB |
| references/effects-and-recovery.md | markdown | 1.7 KB |
| references/generated-program-bindings.md | markdown | 21.1 KB |
| references/programs.md | markdown | 1.5 KB |
references/programs.md
View on GitHub# Programs and exact contracts
Read this before writing `freeflow_run` code. Declare every operation by exact `{id, revision}` in the outer `operations` allowlist; declare captures in `captures`. Inside the restricted JavaScript guest, call `await tools.invoke(id, args)` for operations, `await results.read(id, range)` for allowed captures, and `emit(value)` only for the bounded facts the next model decision needs. Return values that are not emitted remain hidden.
The [generated core bindings](generated-program-bindings.md) describe inputs and canonical values for built-ins. Load the declarations for the selected revisions before authoring their calls. Long-tail declarations come from a complete `freeflow_tools.describe` result. Do not infer a return shape from a model view, add fields from the host's status/coverage/effect envelope to a guest value, or write compatibility parsing for undocumented variants. A guest error exposes `code` and `effectState`; `needs-model` interrupts the run. The host records child coverage, status and partial receipts separately.
Use `mapLimit(items, limit, callback)` for independent bounded reads whose paths and transformation are already known. Sequence dependent operations and mutations. If a child changes what the next step should be, stop the program and let the model decide. Await all calls and emit only the selected evidence; detached calls are not a successful shortcut. Guest code has no ambient Node filesystem, network, modules, clock or random APIs and is not an OS sandbox.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |