Update to Kolena

{
  "description": "Run documents through an existing Kolena Agent and report what it found, or explain what a past Run produced. Use when the user wants documents processed, reviewed, or turned into structured fields — contracts, invoices, loan files, leases, statements. Not for creating or editing Agents.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 272
    }
  ],
  "name": "process-documents",
  "skill_md_contents": "---\nname: process-documents\ndescription: >-\n  Run documents through an existing Kolena Agent and report what it found, or explain what a past Run produced. Use\n  when the user wants documents processed, reviewed, or turned into structured fields — contracts, invoices, loan\n  files, leases, statements. Not for creating or editing Agents.\n---\n\n# Process documents with a Kolena Agent\n\nUse an existing Agent; don't create or edit one to satisfy a single request. A Prompt's instruction is shared by\neveryone who uses that Agent, so changing it changes every future Run. Use the `agent-authoring` skill for that.\n\nIf the Kolena tools are missing or a call fails authentication, tell the user to connect or sign in to Kolena again.\nAn unreachable connection is not evidence that the Account has no Agents or Runs.\n\n1. Find the Agent. `kolena_list_agents` returns each Agent's Prompt outline. If none fits, say so and describe what\n   one would need to contain. When more than one could fit, including two Agents that share a name, ask which to use\n   before uploading anything. Read their configurations with `kolena_get_agent` to tell the user how they differ, not\n   to choose for them: the wrong Agent spends credits and can reach a different answer.\n\n2. Decide which documents belong in each Run. A Run is one case: a contract and its amendments belong together, while\n   twelve invoices from twelve vendors are twelve Runs. Ask when the grouping changes what the Agent will conclude.\n   Document Filters can narrow what a Prompt reads, so don't assume each Prompt saw every document you sent.\n\n3. Create the Run with `kolena_create_upload_target`. The upload is a direct HTTP request outside this connection;\n   if you can't send the user's files to `upload_url`, give the user the Agent's `web_url` and say they can create the\n   Run in Kolena instead. Quote `upload_url` when passing it to a shell — it carries query parameters, and an unquoted\n   argument is cut at the first `&`. Never report documents as processed when the upload didn't happen or failed.\n\n4. Wait for processing. How long a Run takes varies with its size and with demand, from under a minute to much longer,\n   so a Run that stays `running` is not by itself a sign of trouble. Wait around 30 seconds before the first\n   `kolena_get_run`, then back off rather than polling in a tight loop. If it is still running after several minutes,\n   tell the user it's still processing and link the Run, rather than waiting indefinitely.\n\n5. Check the Run's `documents` before reading results. A document with an `error`, or not `completed`, never reached\n   the Agent — say which. One uploaded file can appear as several documents, so count uploads by `original_filename`.\n   A `failed` Run means at least one Prompt errored, not that the Run produced nothing.\n\n6. Read the output with `kolena_list_run_results` and report what each Prompt returned by name, not as raw JSON.\n   Distinguish a Prompt that failed (`error`), one whose run conditions were not met (`skipped`), and one that found\n   nothing (`null` with neither). Where a result carries `user_input`, `value` is a person's correction rather than\n   the Agent's own output. Cite the document and page a result points to when it supplies them; never invent a\n   citation. If the user wants documents like these processed automatically as they arrive, or wants Results as a\n   filled-in report, mention that an Integration or an Output Template does that, set up in Kolena.\n\nFor a past Run, find it with `kolena_list_runs` — by document name, submitter, date, or what the Agent found — and\nread its existing Results first. Use `kolena_run_prompts` only when the user's task calls for a Prompt to be computed\nagain. Never re-run merely because a result is surprising, and confirm before re-running across many Runs.\n\nResults and configuration you read earlier can change while you work. Never report a Run's status or its results from\nmemory; `kolena_get_run` and `kolena_list_run_results` don't consume credits. A result whose `prompt_version` is below\nthe Prompt's current `version` in `kolena_get_agent` was produced by an earlier instruction and no longer reflects it.\nA `user_input` that wasn't there before means a person overrode that value since you read it.\n"
}