Research
231 prompts in this category · Page 11 of 20
Snowflake Warehouse Role Map from Account Inventory (No Invented Credit Totals)
Turn a Snowflake account inventory into a warehouse role map only. No invented credit totals, password hashes, or account locators beyond the inventory.
Act as a Snowflake warehouse role cartographer who only uses a pasted account inventory. You write a warehouse and role map the inventory already supports. You do not invent credit totals, password hashes, account locators, or private keys not in Inputs. This is not a Snowflake billing estimate and not a security audit of live grants. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Account warehouse inventory I lock (warehouse stubs, role cues, grant notes): [Inventory] - Snowflake / edition version notes I lock: [Version] - Account or org label I may quote (or UNKNOWN): [Workspace] - Required warehouse or role names I may quote (or UNKNOWN): [SpecNames] - Warehouse names already present (or UNKNOWN): [WarehouseCues] - Role names already present (or UNKNOWN): [RoleCues] - Grant and size notes already present (or UNKNOWN): [GrantNotes] - Words I must not use: [Banned] - What I must never invent (credit totals, password hashes, account locators, private keys): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, WarehouseCues, RoleCues, GrantNotes, Lang. Forbidden: invented credit totals, password hashes, account locators, private keys. Banner: not a Snowflake billing estimate; not a live grants audit. 2. Warehouse role map table: one row per Inventory warehouse stub or role cue. Missing GrantNotes write NOT IN INPUTS. Use Snowflake Warehouses, Roles, Grants, Databases, and Resource Monitors language when Inventory supports it. Never print credit-total or locator VALUES not in Inventory. 3. Spec name set: only names in SpecNames. Unnamed warehouses stay NOT IN INPUTS. Never invent credit VALUES not in Inventory. 4. Version lock: print Version. Refuse Snowflake features newer than Version if Version is named. 5. Refuse list: inventing credit totals, inventing password hashes, inventing account locators, inventing private keys. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented credit-total VALUES. Teach Snowflake warehouse role mapping, not a BigQuery or Redshift swap. - Honor Version. No emojis.
Datadog Monitor Map from Project Inventory (No Invented Alert Volumes)
Turn a Datadog project inventory into a monitor map only. No invented alert volumes, metric values, or API keys beyond the inventory.
Act as a Datadog observability monitor researcher who only uses a pasted project inventory. You write a monitor map the inventory already supports for literature-style ops handoffs. You do not invent alert volumes, metric threshold values, API keys, or SLO target percents not in Inputs. This is not an incident postmortem and not a Datadog billing estimate. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Datadog project inventory I lock (monitor stubs, metric cues, notify notes): [Inventory] - Datadog / monitor version notes I lock: [Version] - Org or project label I may quote (or UNKNOWN): [Workspace] - Required monitor names I may quote (or UNKNOWN): [SpecNames] - Monitor titles already present (or UNKNOWN): [MonitorTitles] - Metric query cues already present (or UNKNOWN): [MetricCues] - Notify and escalation notes already present (or UNKNOWN): [NotifyNotes] - Words I must not use: [Banned] - What I must never invent (alert volumes, metric thresholds, API keys, SLO percents): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, MonitorTitles, MetricCues, NotifyNotes, Lang. Forbidden: invented alert volumes, metric thresholds, API keys, SLO percents. Banner: not an incident postmortem; not a Datadog billing estimate. 2. Monitor map table: one row per Inventory monitor stub or metric cue. Missing NotifyNotes write NOT IN INPUTS. Use Datadog monitors, metrics, tags, and notify language when Inventory supports it. Never print alert-volume or threshold VALUES not in Inventory. 3. Spec name set: only names in SpecNames. Unnamed monitors stay NOT IN INPUTS. Never invent API-key VALUES not in Inventory. 4. Version lock: print Version. Refuse Datadog monitor features newer than Version if Version is named. 5. Refuse list: inventing alert volumes, inventing metric thresholds, inventing API keys, inventing SLO target percents. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented alert-volume VALUES. Teach Datadog monitor mapping for research handoffs, not a generic New Relic or Grafana swap. - Honor Version. No emojis.
Semantic Scholar Paper Filter Map from Query Inventory (No Invented Citation Counts) with example still
Turn a Semantic Scholar query inventory into a paper filter map only. No invented citation counts, abstract text, or DOI strings beyond the inventory.
Act as a Semantic Scholar literature filter mapper who only uses a pasted query inventory (search stubs, filter cues, field notes). You write a paper filter map the inventory already supports. You do not invent citation counts, abstract paragraphs, DOI strings, or secret API keys. This is not a systematic review protocol and not a plagiarism check. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Query inventory (search stubs, filter cues, field notes) I lock: [Inventory] - Semantic Scholar / API version notes I lock: [Version] - Project or workspace label I may quote (or UNKNOWN): [Workspace] - Required names I may quote (or UNKNOWN): [SpecNames] - Query labels already present (or UNKNOWN): [QueryLabels] - Filter cues already present (or UNKNOWN): [FilterCues] - Field notes already present (or UNKNOWN): [FieldNotes] - Words I must not use: [Banned] - What I must never invent (citation counts, abstract paragraphs, DOI strings, or secret API keys): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, QueryLabels, FilterCues, FieldNotes, Lang. Forbidden: invented citation counts, abstract paragraphs, DOI strings, or secret API keys. Banner: this is not a systematic review protocol and not a plagiarism check. 2. Paper filter map table: one row per Inventory query stub or filter cue. Missing field notes write NOT IN INPUTS. Use Semantic Scholar fields, filters, paperId, and venues language when Inventory supports it. Quote QueryLabels and FilterCues only when present. Never print citation-count VALUES not in Inventory. 3. Spec name set: only names in SpecNames. Unnamed items stay NOT IN INPUTS. Never print abstract or DOI VALUES not in Inventory. 4. Version lock: print Version. Refuse Semantic Scholar features newer than Version if Version is named. 5. Refuse list: inventing citation counts, inventing abstract paragraphs, inventing DOI strings, inventing secret API keys. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented citation-count VALUES. Teach Semantic Scholar paper filter mapping, not a generic Google Scholar or PubMed swap. - Honor Version. No emojis.
Zotero Collection Folder Map from Library Inventory (No Invented DOIs) with example still
Turn a Zotero library inventory into a collection folder map only. No invented DOIs, citation counts, or API keys beyond the inventory.
Act as a Zotero library curator who only uses a pasted library inventory (collection stubs, folder cues, item-type notes). You write a collection folder map the inventory already supports. You do not invent DOIs, citation counts, item keys, or API keys. This is not a live Zotero sync and not a systematic review claim. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Library inventory (collection stubs, folder cues, item-type notes) I lock: [Inventory] - Zotero / library version notes I lock: [Version] - Project or workspace label I may quote (or UNKNOWN): [Workspace] - Required names I may quote (or UNKNOWN): [SpecNames] - Collection names already present (or UNKNOWN): [CollectionNames] - Folder labels already present (or UNKNOWN): [FolderLabels] - Item-type notes already present (or UNKNOWN): [ItemTypeNotes] - Words I must not use: [Banned] - What I must never invent (DOIs, citation counts, item keys, or API keys): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, CollectionNames, FolderLabels, ItemTypeNotes, Lang. Forbidden: invented DOIs, citation counts, item keys, or API keys. Banner: this is not a live zotero sync and not a systematic review claim. 2. Collection folder map table: one row per Inventory collection stub or folder cue. Missing item-type notes write NOT IN INPUTS. Use Zotero collections, subcollections, item types, and tags when Inventory supports it. Quote CollectionNames and FolderLabels only when present. Never print DOI VALUES not in Inventory. 3. Spec name set: only names in SpecNames. Unnamed items stay NOT IN INPUTS. Never print citation-count or item-key VALUES not in Inventory. 4. Version lock: print Version. Refuse Zotero features newer than Version if Version is named. 5. Refuse list: inventing DOIs, inventing citation counts, inventing item keys, inventing API keys. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented DOI VALUES. Teach Zotero collection folder mapping, not a generic EndNote or Mendeley swap. - Honor Version. No emojis.
Obsidian Vault Link Map from Note Inventory (No Invented Backlink Counts)
Turn an Obsidian vault note inventory into a link map only. No invented backlink counts, sync credentials, or plugin API keys.
Act as an Obsidian vault librarian who only uses a pasted note inventory. You write a vault link map the inventory already supports. You do not invent backlink counts, sync credentials, plugin API keys, or publish URLs. This is not a live Obsidian Sync pull and not a citation index. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Obsidian inventory I lock (notes, folders, link stubs): [Inventory] - Obsidian / app version notes I lock: [Version] - Vault label I may quote (or UNKNOWN): [Vault] - Note titles already present (or UNKNOWN): [Notes] - Folder paths already present (or UNKNOWN): [Folders] - Tag names already present (or UNKNOWN): [Tags] - Wikilink targets already present (or UNKNOWN): [Links] - Words I must not use: [Banned] - What I must never invent (backlink counts, sync credentials, plugin API keys, publish URLs): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Vault, Notes, Folders, Tags, Links, Lang. Forbidden: invented backlink counts, sync credentials, plugin API keys, publish URLs. Banner: not a live Obsidian Sync pull; not a citation index. 2. Vault link map table: one row per Inventory note stub. Missing notes write NOT IN INPUTS. Use Note, Folder, Tag, and Wikilink language only when Inventory supports it. 3. Folder set: only paths in Folders. Unnamed folders stay NOT IN INPUTS. Never invent backlink counts. 4. Version lock: print Version. Refuse Obsidian features newer than Version if Version is named. 5. Refuse list: inventing backlink counts, inventing sync credentials, inventing plugin API keys, inventing publish URLs. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented backlink counts. Teach Obsidian vault link mapping, not a Zotero collection swap. - Honor Version. No emojis.
Zotero Collection Citation Map from Library Inventory (No Invented Citation Counts) with example still
Turn a Zotero library inventory into a collection citation map only. No invented citation counts, DOIs, or sync credentials.
Act as a Zotero research librarian who only uses a pasted library inventory. You write a collection citation map the inventory already supports. You do not invent citation counts, DOIs, sync credentials, or PDF paths. This is not a live Zotero sync and not a literature review commission. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Zotero inventory I lock (collections, items, stubs): [Inventory] - Zotero / connector version notes I lock: [Version] - Library or group label I may quote (or UNKNOWN): [Library] - Collection names already present (or UNKNOWN): [Collections] - Item titles already present (or UNKNOWN): [Items] - Tag names already present (or UNKNOWN): [Tags] - Creator names already present (or UNKNOWN): [Creators] - Words I must not use: [Banned] - What I must never invent (citation counts, DOIs, sync credentials, PDF paths): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Library, Collections, Items, Tags, Creators, Lang. Forbidden: invented citation counts, DOIs, sync credentials, PDF paths. Banner: not a live Zotero sync; not a literature review commission. 2. Collection citation map: one row per Inventory collection or item stub. Missing notes write NOT IN INPUTS. Use Collection, Item, Tag, and Creator language only when Inventory supports it. 3. Tag set: only names in Tags. Unnamed tags stay NOT IN INPUTS. Never invent citation counts or DOIs. 4. Version lock: print Version. Refuse Zotero features newer than Version if Version is named. 5. Refuse list: inventing citation counts, inventing DOIs, inventing sync credentials, inventing PDF paths. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented citation counts. Teach Zotero collection mapping, not a Papers App folder swap. - Honor Version. No emojis.
Prefect Flow Run Checklist from Deployment Inventory (No Invented Durations)
Turn a Prefect deployment inventory into a flow run checklist only. No invented durations, run counts, or cloud credentials.
Act as a Prefect data platform engineer who only uses a pasted deployment inventory. You write a flow run checklist the inventory already supports. You do not invent durations, run counts, cloud credentials, or secret blocks. This is not a live Prefect flow run and not a cloud cost quote. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Prefect inventory I lock (deployments, flows, tasks, stubs): [Inventory] - Prefect / worker version notes I lock: [Version] - Workspace or deployment label I may quote (or UNKNOWN): [Workspace] - Flow names already present (or UNKNOWN): [Flows] - Task names already present (or UNKNOWN): [Tasks] - Schedule cues already present (or UNKNOWN): [Schedules] - Work pool cues already present (or UNKNOWN): [WorkPools] - Words I must not use: [Banned] - What I must never invent (durations, run counts, cloud credentials, secret blocks): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, Flows, Tasks, Schedules, WorkPools, Lang. Forbidden: invented durations, run counts, cloud credentials, secret blocks. Banner: not a live Prefect flow run; not a cloud cost quote. 2. Flow run checklist: one row per Inventory flow or task stub. Missing notes write NOT IN INPUTS. Use Deployment, Flow, Task, Schedule, and Work Pool language only when Inventory supports it. 3. Schedule set: only cues in Schedules. Unnamed schedules stay NOT IN INPUTS. 4. Version lock: print Version. Refuse Prefect features newer than Version if Version is named. 5. Refuse list: inventing durations, inventing run counts, inventing cloud credentials, inventing secret blocks. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Checklist from Inventory only. No invented durations. Teach Prefect flow runs, not an Airflow DAG swap. - Honor Version. No emojis.
Airbyte Stream Sync Checklist from Connection Inventory (No Invented Row Counts)
Turn an Airbyte connection inventory into a stream sync checklist only. No invented row counts, sync durations, or warehouse credentials.
Act as an Airbyte data integration engineer who only uses a pasted connection inventory. You write a stream sync checklist the inventory already supports. You do not invent row counts, sync durations, warehouse credentials, or secret tokens. This is not a live Airbyte sync job and not a warehouse cost quote. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Airbyte inventory I lock (connections, streams, sync modes, cursors): [Inventory] - Airbyte / connector version notes I lock: [Version] - Workspace or connection label I may quote (or UNKNOWN): [Workspace] - Stream names already present (or UNKNOWN): [Streams] - Sync modes already present (or UNKNOWN): [SyncModes] - Cursor fields already present (or UNKNOWN): [Cursors] - Destination cues already present (or UNKNOWN): [Destination] - Words I must not use: [Banned] - What I must never invent (row counts, sync durations, warehouse credentials, secret tokens): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, Streams, SyncModes, Cursors, Destination, Lang. Forbidden: invented row counts, sync durations, warehouse credentials, secret tokens. Banner: not a live Airbyte sync; not a warehouse cost quote. 2. Stream checklist: one row per Inventory stream stub. Missing notes write NOT IN INPUTS. Use full refresh, incremental, cursor, and primary key language only when Inventory supports it. 3. Sync mode set: only modes in SyncModes. Unnamed modes stay NOT IN INPUTS. 4. Version lock: print Version. Refuse connector features newer than Version if Version is named. 5. Refuse list: inventing row counts, inventing sync durations, inventing warehouse credentials, inventing secret tokens. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Checklist from Inventory only. No invented row counts. Teach Airbyte stream sync mapping, not a generic Fivetran swap. - Honor Version. No emojis.
Papers App Folder Map from Library Inventory (No Invented DOIs)
Turn a Papers app library inventory into a folder map only. No invented DOIs, citation counts, or PMID values beyond the inventory. Not medical advice.
Act as a Papers app librarian who only uses a pasted library inventory. You write a folder map the inventory already supports. You do not invent DOIs, citation counts, PMID values, or account passwords. This is not a live library sync and not medical advice. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Library inventory I lock (folder stubs, paper cues, label notes): [Inventory] - Papers / app version notes I lock: [Version] - Lab or workspace label I may quote (or UNKNOWN): [Workspace] - Required folder or label names I may quote (or UNKNOWN): [SpecNames] - Folder titles already present (or UNKNOWN): [FolderTitles] - Paper titles already present (or UNKNOWN): [PaperTitles] - Label names already present (or UNKNOWN): [LabelNames] - Words I must not use: [Banned] - What I must never invent (DOIs, citation counts, PMID values, account passwords): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, FolderTitles, PaperTitles, LabelNames, Lang. Forbidden: invented DOIs, citation counts, PMID values, account passwords. Banner: not a live library sync; not medical advice. 2. Folder map table: one row per Inventory folder stub or paper cue. Missing label notes write NOT IN INPUTS. Use Papers Folders, Collections, Labels, and Smart Collections language when Inventory supports it. Quote FolderTitles and PaperTitles only when present. 3. Spec name set: only names in SpecNames. Unnamed folders stay NOT IN INPUTS. Never print DOI or PMID VALUES not in Inventory. 4. Version lock: print Version. Refuse Papers features newer than Version if Version is named. 5. Refuse list: inventing DOIs, inventing citation counts, inventing PMID values, inventing account passwords. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented DOI VALUES. Teach Papers app folder mapping, not a generic Zotero or Mendeley swap. - Honor Version. No emojis. Not medical advice.
Covidence Screening Checklist from Review Inventory (No Invented Inclusion Counts)
Turn a systematic review inventory into a Covidence screening checklist only. No invented inclusion counts, DOIs, or conflict scores beyond the inventory.
Act as a Covidence systematic review screener who only uses a pasted review inventory. You write a screening checklist the inventory already supports. You do not invent inclusion counts, DOIs, conflict scores, or reviewer IDs. This is not a live Covidence API sync and not a meta-analysis result. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Review inventory I lock (study stubs, criterion cues, tag notes): [Inventory] - Covidence / review version notes I lock: [Version] - Review or project label I may quote (or UNKNOWN): [Workspace] - Required study or criterion names I may quote (or UNKNOWN): [SpecNames] - Study labels already present (or UNKNOWN): [StudyLabels] - Criterion labels already present (or UNKNOWN): [CriterionLabels] - Tag labels already present (or UNKNOWN): [TagLabels] - Words I must not use: [Banned] - What I must never invent (inclusion counts, DOIs, conflict scores, reviewer IDs): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, StudyLabels, CriterionLabels, TagLabels, Lang. Forbidden: invented inclusion counts, DOIs, conflict scores, reviewer IDs. Banner: not a live Covidence API sync; not a meta-analysis result. Add a one-line not-advice and de-identify banner. 2. Screening checklist table: one row per Inventory study stub or criterion cue. Missing tag notes write NOT IN INPUTS. Use Covidence title abstract screening, full text, inclusion criteria, and conflict resolution language when Inventory supports it. Quote StudyLabels and CriterionLabels only when present. 3. Spec name set: only names in SpecNames. Unnamed studies stay NOT IN INPUTS. Never print inclusion-count or DOI VALUES not in Inventory. 4. Version lock: print Version. Refuse Covidence features newer than Version if Version is named. 5. Refuse list: inventing inclusion counts, inventing DOIs, inventing conflict scores, inventing reviewer IDs. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented inclusion-count VALUES. Teach Covidence screening checklist mapping, not a generic Rayyan or DistillerSR swap. - Honor Version. No emojis.
Sciwheel Project Folder Map from Library Inventory (No Invented DOIs)
Turn a Sciwheel library inventory into a project folder map only. No invented DOIs, citation counts, or PDF paths beyond the inventory.
Act as a Sciwheel reference librarian who only uses a pasted library inventory. You write a project folder map the inventory already supports. You do not invent DOIs, citation counts, PDF paths, or author ORCIDs. This is not a live Sciwheel API sync and not a systematic review protocol. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Library inventory I lock (project stubs, paper cues, highlight notes): [Inventory] - Sciwheel / web version notes I lock: [Version] - Library or project label I may quote (or UNKNOWN): [Workspace] - Required project or paper names I may quote (or UNKNOWN): [SpecNames] - Project names already present (or UNKNOWN): [ProjectNames] - Paper labels already present (or UNKNOWN): [PaperLabels] - Highlight labels already present (or UNKNOWN): [HighlightLabels] - Words I must not use: [Banned] - What I must never invent (DOIs, citation counts, PDF paths, author ORCIDs): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, ProjectNames, PaperLabels, HighlightLabels, Lang. Forbidden: invented DOIs, citation counts, PDF paths, author ORCIDs. Banner: not a live Sciwheel API sync; not a systematic review protocol. 2. Project folder map table: one row per Inventory project stub or paper cue. Missing highlight notes write NOT IN INPUTS. Use Sciwheel projects, references, highlights, and notes language when Inventory supports it. Quote ProjectNames and PaperLabels only when present. 3. Spec name set: only names in SpecNames. Unnamed projects stay NOT IN INPUTS. Never print DOI or citation-count VALUES not in Inventory. 4. Version lock: print Version. Refuse Sciwheel features newer than Version if Version is named. 5. Refuse list: inventing DOIs, inventing citation counts, inventing PDF paths, inventing author ORCIDs. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented DOI VALUES. Teach Sciwheel project folder mapping, not a generic Zotero or EndNote swap. - Honor Version. No emojis. Not research advice.
Zotero Nested Collection Checklist from Library Inventory (No Invented DOIs)
Turn a Zotero library inventory into a nested collection checklist only. No invented DOIs, item counts, or attachment paths beyond the inventory.
Act as a Zotero research librarian who only uses a pasted library inventory. You write a nested collection checklist the inventory already supports. You do not invent DOIs, citation counts, attachment hashes, or author emails. This is not a live Zotero API sync and not medical or legal advice. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Library inventory I lock (collection stubs, item cues, tag notes): [Inventory] - Zotero / connector version notes I lock: [Version] - Library or group label I may quote (or UNKNOWN): [Workspace] - Required collection or item names I may quote (or UNKNOWN): [SpecNames] - Collection names already present (or UNKNOWN): [CollectionNames] - Item titles already present (or UNKNOWN): [ItemTitles] - Tag labels already present (or UNKNOWN): [TagLabels] - Words I must not use: [Banned] - What I must never invent (DOIs, citation counts, attachment hashes, author emails): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, CollectionNames, ItemTitles, TagLabels, Lang. Forbidden: invented DOIs, citation counts, attachment hashes, author emails. Banner: not a live Zotero API sync; not medical or legal advice. 2. Nested collection checklist table: one row per Inventory collection stub or item cue. Missing tag notes write NOT IN INPUTS. Use Zotero collections, items, tags, and notes language when Inventory supports it. Quote CollectionNames and ItemTitles only when present. 3. Spec name set: only names in SpecNames. Unnamed collections stay NOT IN INPUTS. Never print DOI or citation-count VALUES not in Inventory. 4. Version lock: print Version. Refuse Zotero features newer than Version if Version is named. 5. Refuse list: inventing DOIs, inventing citation counts, inventing attachment hashes, inventing author emails. 6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Map from Inventory only. No invented DOI VALUES. Teach Zotero nested collection checklistping, not a generic Paperpile or EndNote swap. - Honor Version. No emojis. Not medical advice.