enterprise-information-search-scope
Approve the selected working set of enterprise artifacts and question-to-source mappings needed to answer the question file.
- License 许可
- MIT
- 最近更新
- 2026-09-30
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- enterprise-information-search-scope
- description
- Approve the selected working set of enterprise artifacts and question-to-source mappings needed to answer the question file.
03 SKILL.md 正文
Enterprise Information Search Scope
Use this stage after workflow/question_intake_checkpoint.json exists and before deep retrieval or answer writing. The enterprise dataset mixes metadata files and per-product artifact bundles, so this stage approves a narrow, reviewable search scope, records which enterprise artifacts are selected for each question key, and leaves the work in pending continuation status instead of treating the task as complete.
Inputs
workflow/question_intake_checkpoint.json/root/DATA
Use the question keys, target entities, and artifact clues already frozen in workflow/question_intake_checkpoint.json as the source of truth for this stage. Do not broaden the task by rebuilding intent from scratch when the checkpoint already captured it.
Selected Working Set Approval
- Read
workflow/question_intake_checkpoint.jsonand extract the question keys that still need answers, the target entities each question asks for, and any explicit product, document, channel, meeting, PR, URL, employee, or customer clues. - Survey
/root/DATAat a bounded level to find candidate source files.- Use metadata files to disambiguate names, employee IDs, customer IDs, roles, orgs, and team structure.
- Use product grounding before approving any product bundle. A product file should only enter the selected working set when the checkpoint or corroborating artifact clues tie it to one or more question keys.
- If the checkpoint names or points to an operational note or handoff clue, keep that path eligible for later route binding instead of discarding it early.
- Build
selected_candidatesandnon_selected_candidatesexplicitly.- Every
selected_candidatesentry must have a concrete role in answering at least one question key. - Every
non_selected_candidatesentry must record a short, specificwhy_excludedreason. - Separate similar product names, aliases, or report variants using metadata and product grounding before approval. If grounding is not established, exclude the candidate for now instead of forcing it into scope.
- Every
- Leave the approved working set pending continuation. Do not mark the task complete in this stage.
Question-to-Source Mappings
- Build
question_source_mapfrom the approved working set. - Map each question key such as
q1to the exact source paths fromselected_candidatesthat should be searched next. - Use the same path strings in
question_source_mapthat appear inselected_candidates. - Prefer the narrowest path that still covers the question. A specific metadata file or product bundle path is better than a broad directory when the scope is already known.
- The same path may appear more than once in
selected_candidatesif differentartifact_typevalues or differentquestion_keysjustify distinct downstream use.
Search Scope Summary
Write workflow/search_scope_summary.json with the exact question keys and the downstream rules that keep answer formatting consistent:
question_keystarget_entitiespreferred_artifact_typesanswer_list_ruletoken_rule
Use this exact preferred_artifact_types order:
[
"metadata",
"documents",
"slack",
"meeting_transcripts",
"meeting_chats",
"prs",
"urls"
]
Set these exact string values:
answer_list_rule:every answer stays a listtoken_rule:every question records consumed tokens as a positive number
Output Contract
workflow/approved_answer_working_set.json
{
"selected_candidates": [
{
"path": "string",
"artifact_type": "string",
"question_keys": ["string"],
"why_selected": "string"
}
],
"non_selected_candidates": [
{
"path": "string",
"why_excluded": "string"
}
],
"question_source_map": {
"q1": ["string"]
},
"pending_continuation_status": "approved-not-final"
}
workflow/search_scope_summary.json
{
"question_keys": ["string"],
"target_entities": ["string"],
"preferred_artifact_types": [
"metadata",
"documents",
"slack",
"meeting_transcripts",
"meeting_chats",
"prs",
"urls"
],
"answer_list_rule": "every answer stays a list",
"token_rule": "every question records consumed tokens as a positive number"
}
Minimal Validation
Before handoff:
- Confirm every question key from the checkpoint appears in
question_source_mapandquestion_keys. - Confirm every
selected_candidatesentry lists at least one question key. - Confirm
pending_continuation_statusis exactlyapproved-not-final. - Confirm
preferred_artifact_types,answer_list_rule, andtoken_rulematch the contract exactly. - Confirm any operational note or handoff clue referenced by the checkpoint was not discarded into
non_selected_candidatesbefore later continuation can use it.
Continuation Status
The next stage must consume these exact keys without renaming them: selected_candidates, non_selected_candidates, question_source_map, pending_continuation_status, question_keys, target_entities, preferred_artifact_types, answer_list_rule, and token_rule.
Treat workflow/approved_answer_working_set.json and workflow/search_scope_summary.json as the canonical working record for deeper retrieval. Stop after both workflow files are written and validated. Do not write /root/answer.json in this stage.
Routing-Depth Continuation
After this stage's local outputs are complete, read the enterprise-information-search-binder skill before continuing. This is a routing instruction only: do not substitute direct task completion for the designated successor's work.
04 同仓库更多 Skills
video-silence-remover-scope-verifier
Use only when `workflow/video-silence-remover-board.json` selects video-silence-remover-scope-verifier and carries a matching continuation result and lease. Verify exactly that task mutation; not a startup, repair, or task-planning entrypoint.
video-silence-remover-scope-binder
Use only when `workflow/video-silence-remover-board.json` explicitly selects video-silence-remover-scope-binder and names one active task item. Bind exactly one epoch-scoped mutation lease; not a startup, discovery, or task-planning entrypoint.
video-silence-remover-entry-selector
Use first for the entire requested task. Before broad task discovery, source edits, global verification, a user-task continuation, or a summary, create or read `workflow/video-silence-remover-board.json`; use this entry whenever that board is absent or selects this entry after the return gate. This is the only startup-authoritative ring stage and mandatory return selector: every return must consume its checkpoint and publish one packet before any other task work. Reading it commits a board read; an existing return then commits ticket consumption, one fixed-path probe, and the complete binder handoff. The persistent board forbids cleanup and final response.
video-silence-remover-catalog-return-gate
Use only when `workflow/video-silence-remover-board.json` selects video-silence-remover-catalog-return-gate and carries one scoped verifier status. Reading this skill commits an immediate board read followed by one board-only gate transition and its marker; remaining benchmark work cannot be selected here.
video-silence-remover-authorized-continuation
Use only when `workflow/video-silence-remover-board.json` selects video-silence-remover-authorized-continuation and carries its matching active lease. Perform exactly one leased task mutation; not a startup, discovery, verification, or task-planning entrypoint.
threejs-to-obj-scope-verifier
Use only when `workflow/threejs-to-obj-board.json` selects threejs-to-obj-scope-verifier and carries a matching continuation result and lease. Verify exactly that task mutation, then immediately hand its still-unaccepted receipt to the return gate; not a startup, repair, task-planning, or final-acceptance entrypoint.