القائمة
Project already-captured traces into scored-rollout rows for the SkillOpt optimizer.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Share agent traces to open datasets on HuggingFace Hub. Use this skill when the user mentions OpenTraces, trace capture, Trace Trails, workflow-built datasets, dataset review, or publishing reviewed dataset rows.
Share agent traces to open datasets on HuggingFace Hub. Use this skill when the user mentions OpenTraces, trace capture, Trace Trails, workflow-built datasets, dataset review, or publishing reviewed dataset rows.
Per-commit trace lineage rows for a branch context, deterministic.
Derive compact eval rows from command-attributed trace trajectories.
Land every site-related fix on main, deploy opentraces.ai to Vercel production, and verify it works across all pages (desktop + mobile) and the email-capture flow, fixing and re-deploying until green. Runs as a goal: land -> deploy -> verify -> triage -> repeat until safe. Use when the user says "deploy site", "deploy to vercel", "push site", "ship the site", "deploy-site", or "land and verify the site". For a full coordinated release that also bumps the version and publishes the Python packages, use /release-pack instead.
Take one or more GitHub issues and drive each to a review-ready pull request, holding one belief throughout: an issue is a hypothesis, not a spec. The pipeline VERIFIES BEFORE IT TRUSTS and runs TWO ADVERSARIES before it commits: multi-agent investigation of the real cause (flagging any "landmine" — a cited function or file:line that does not exist) and the blast radius, a plan whose test must go RED BEFORE GREEN, two adversarial gates on that plan (a self/ethos checklist + a pure codex adversarial pass), implementation expressed as a forged `/goal` (via goal-forge) whose verify loop runs the otbox acceptance journey, a green suite + docs sync, a PR that `Closes #<n>`, and a FINAL codex adversarial review of the diff. Honesty, not theater: it reports unmet acceptance criteria and invalid issues rather than faking a fix. The human merges. Everything happens in a per-issue git worktree; multiple issues each get their own worktree + branch + PR. Use when the user says "issue-to-pr", "solve issue", "fix issue #N"
| name | skill-opt-v1 |
| description | Project already-captured traces into scored-rollout rows for the SkillOpt optimizer. |
| mode | agent-skill |
| requires | ["ot workflow optimize","ot trace query"] |
Scored-rollout projection for the SkillOpt text-space skill optimizer
(arXiv 2605.23904; design in kb/br/66-skillopt-text-space-skill-optimizer.md).
scripts/build_rows.py reads each already-captured trace from the local bucket,
scores it with opentraces.quality.engine.assess_trace, and emits one row per
trace. The trace's overall_utility (in [0,1] after scaling) is the rollout
score; the failed and passed quality checks become failure_tags and
success_tags, which the optimizer's reflection step and the held-out
validation gate consume.
This workflow is the forward-pass-over-captured-evidence layer. It does not run
an agent; the live re-rollout of a candidate skill is a later slice. The
opentraces workflow optimize consumer reads these rows, splits them into a
train and a held-out selection set by trace-id hash, proposes bounded
add/delete/replace edits, accepts only strictly-improving candidates, and
exports best_skill.md plus edit_apply_report.json.
{
"trace_id": "…",
"project_slug": "…",
"session_id": "…",
"score": 0.62, // overall_utility / 100, the rollout score
"overall_utility": 62.0,
"failure_tags": ["rl.RL1_grounded_outcome", "training.T3_clean_diff"],
"success_tags": ["conformance.C1_schema_version"],
"summary": "task description (truncated)",
"skill_digest": null
}
The run packet's scope may carry project_slug (or project) to narrow the
bucket scan to one project; otherwise every project's traces are projected.