Skip to content

RFC 0022: dataarc recipe for terminal_synth

Status: experimental implementation in PR #109; release evidence below Author: @adithya-s-k Created: 2026-09-11

Summary

Use the released terminal synthesis workflow to create an instruction, assets, solution and verifier, followed by execution-informed revision. Limit claims to the released terminal branch and enforce the concrete generated APIs.

Motivation

Bring the released method into Repo2RLEnv as owned, maintainable code, with standalone Harbor output and independent quality evidence. The upstream project remains the attribution and comparison baseline, not a runtime dependency. Shared operations use RFC 0011; method-specific generation decisions stay in this recipe.

Design

Input

Native input: Upstream seed Harbor tasks and supported few/self/evol settings.

Select pipeline.name: terminal_synth and pipeline.recipe: dataarc in a typed configuration. Source data, resolved revisions, resource limits, model roles, random seeds and recipe options are recorded before spending. Strict options reject unknown keys. Existing native pipeline defaults remain compatible.

Algorithm

Use the released terminal synthesis workflow to create an instruction, assets, solution and verifier, followed by execution-informed revision. Limit claims to the released terminal branch and enforce the concrete generated APIs.

Each substantive stage emits a typed progress event and an artifact-bound receipt. Classification separates source eligibility, infrastructure, oracle, verifier and solver failures. Repairs are bounded and invalidate affected downstream evidence.

Output

A complete Harbor bundle: instruction, task configuration, environment, reference entry point and trusted verifier. Metadata records recipe/version, input lineage, upstream source pin, adaptations, reward scale, image/asset digests and the complete task hash. Exported is distinct from accepted.

Verification

The linked pipeline guide specifies the implemented native generation checks. Generation exports and independent quality acceptance are separate: reference success does not establish verifier coverage or shortcut resistance. Shared review, repair and labeling contracts are defined in RFC 0027.

Anti-contamination

Learner-visible snapshots exclude the reference, future Git objects, credentials and private tests. Execute grading so learner code cannot inspect the private oracle. Prefetch pinned assets; enforce and probe the actual learner network policy. The prompt is not an access-control mechanism. Preserve legitimate source context rather than indiscriminately deleting it.

LLM use

Where the algorithm requires synthesis or review, use recorded role-specific models through the common metered client or agent adapter. Reference execution is deterministic. Cost includes failures, retries, bootstrap, cloud runtime and independent audits; unknown costs are not zero.

Yield and suitability

See the pipeline guide for supported inputs and measured economics for sample sizes, yield definitions and cost coverage. Results on a selected source profile do not imply universal input conversion.

Dependencies

Repository-owned recipe code, existing source/auth/LLM/bootstrap helpers, remote execution adapters and Harbor. Essential SDKs and ordinary libraries are permitted. No install/import/clone of the upstream research implementation at runtime. No dependency on ignored local reference folders or private pilot artifacts.

Alternatives considered

A wrapper around upstream commands would preserve an uncontrolled runtime dependency. One generic generator for every method would lose method-specific behavior. Use owned stages with common execution and quality contracts instead; explicitly version deviations from the upstream baseline.

References

Implementation

Owned implementation: pipelines/recipes/dataarc/. Version 2's authored prompt and four strategy instructions feed direct artifact generation over complete Harbor seeds. Typed materialization, full environment context, remote baseline/reference checks and bounded repairs are explicit adaptations. Tests cover strategy enumeration, context filtering, source identity and missing-reference rejection. See the guide and examples/owned-dataarc.yaml. Quality acceptance remains deferred until the generation milestone across all recipes.

Version 1 retained prompt text from the terminal branch and a license from main, which does not contain that implementation. Version 2 removes both retained artifacts. New exports record the recipe revision; runtime-wheel identity prevents resuming an old run with changed prompt resources. Published version 1 tasks are historical results, not evidence of version 2 generation quality or licensing.

Current release evidence

100 published Harbor tasks. The pipeline walkthrough documents the implemented profile, actual model calls, bounded repairs and limitations. The release inventory records source diversity, scoped economics, artifact revisions and evaluation labels. Generation checks, independent review and blind solver success are separate claims.