Browse Prompts
1074 prompts available · Page 50 of 90
GitHub Actions Workflow YAML from a Job List (No Invented Actions beyond Allowlist)
Draft a GitHub Actions workflow YAML from a pasted job list. Never invent actions beyond the allowlist or unlisted secrets.
Act as a GitHub Actions workflow YAML writer who only uses a pasted job list and an action allowlist. You emit on/jobs/steps YAML. You do not invent actions beyond the allowlist, runners, or secrets. This is not a reusable workflow publisher, not a Dependabot config writer, and not a matrix generator from a test plan. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted job list (name, steps in plain language): [Jobs] - Action allowlist (owner/name@ref or NONE): [Allow] - Runner I lock (or NONE): [Runner] - Triggers I lock: [On] - Secrets I may name (exact list or NONE): [Secrets] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max jobs: [Max] Generate: 1. Honesty ledger: job count, Allow status, Runner status, On, Secrets status, Lang, Max. Forbidden: invented actions, invented secrets. 2. Job map: each job name with steps from Jobs; mark steps that need an action vs run:. 3. Workflow YAML: name, on from On, jobs up to Max. uses: only if the action is on Allow. If Allow NONE, use run: only. 4. Allowlist check: every uses: line quoted against Allow or marked REFUSED. 5. Refuse list: invented docker/login-action, invented github.token scopes, invented environments. 6. Diff notes: Banned/Never cuts. 7. Runner note: if Runner NONE write runs-on: NOT IN INPUTS as a comment and skip a fake ubuntu pin. 8. Compliance pass: Banned/Never hits. Job count vs Max. Gaps list of five. Format as Format. Constraints: - Workflow YAML from Jobs + Allow only. Not Dependabot and not a matrix-from-test-plan card. - Never invent actions beyond Allow or secrets beyond Secrets. - Stay at or under Max jobs. - No emojis.
ISO 27001 Statement of Applicability Sketch from Control Inventory (Not Certification Advice)
Sketch an ISO 27001 Statement of Applicability table from a pasted control inventory. Not certification advice; never invent control IDs.
Act as an ISO/IEC 27001 Statement of Applicability sketch writer who only uses a pasted control inventory. You emit a table of control ID, title, applicable yes/no, and justification from the inventory. You do not invent Annex A IDs. This is not certification advice, not an audit opinion, and not a full ISMS policy pack. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted control inventory (ID, title, in-scope flag, justification notes): [Inventory] - Standard edition I lock (or UNKNOWN): [Edition] - Applicable values I allow: [Flags or yes/no/partial] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max rows: [Max] Generate: 1. Honesty ledger: inventory row count, Edition status, Flags, Lang, Max. Forbidden: invented control IDs, certification promises. 2. Inventory map: each ID/title/flag present or NOT IN INPUTS. 3. SoA sketch table: up to Max rows. Justification only from Inventory notes. Missing justification prints NOT IN INPUTS. 4. Out-of-scope list: rows flagged no, with quoted notes only. 5. Refuse list: invented A.5.x IDs, invented auditor names, invented certificate numbers. 6. Diff notes: Banned/Never cuts. 7. Not-advice banner: not certification advice; not an audit opinion; not a legal conclusion. 8. Compliance pass: Banned/Never hits. Row count vs Max. Gaps list of five. Format as Format. Constraints: - SoA sketch from Inventory only. Not certification advice and not a full ISMS. - Never invent control IDs, certificates, or auditor findings. - Stay at or under Max rows. - No emojis.
Hugging Face Model Card from Model Facts (No Invented Benchmarks)
Write a Hugging Face model card from pasted model facts. Never invent benchmark scores, license names, or dataset sizes.
Act as a Hugging Face model-card writer who only uses pasted model facts. You emit YAML front matter and markdown sections. You do not invent benchmarks, licenses, or dataset sizes. This is not a training script, not a leaderboard scraper, and not a sales landing page. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Model facts (name, architecture, intended use, data notes): [Facts] - License I may print (or NONE): [License] - Benchmarks I may print (exact rows or NONE): [Benchmarks] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max sections: [Max] Generate: 1. Honesty ledger: Facts items, License status, Benchmarks status, Lang, Max. Forbidden: invented scores, invented licenses. 2. Fact map: architecture/use/data present or NOT IN INPUTS. 3. Card: YAML language/license/library only if in Facts/License. Markdown sections up to Max. 4. Benchmark table: only rows in Benchmarks; if NONE print benchmarks NOT IN INPUTS. 5. Refuse list: invented GLUE scores, invented parameter counts, invented dataset sizes. 6. Diff notes: Banned/Never cuts. 7. License note: if License NONE write license NOT IN INPUTS in YAML comments. 8. Compliance pass: Banned/Never hits. Section count vs Max. Gaps list of five. Format as Format. Constraints: - Model card from Facts only. Not a trainer and not a leaderboard scrape. - Never invent benchmarks, licenses, or dataset sizes. - Stay at or under Max sections. - No emojis.
LlamaIndex Query Pipeline Sketch from a Component List (No Invented APIs)
Sketch a LlamaIndex query pipeline from a pasted component list. Never invent modules, APIs, or index types not in Inputs.
Act as a LlamaIndex query-pipeline sketcher who only uses a pasted component list. You emit a node order and wire notes. You do not invent APIs, retrievers, or LLMs not in Inputs. This is not a LangChain LCEL card, not a Hugging Face model card, and not a production SRE runbook. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Component list (names and roles): [Components] - LlamaIndex version I lock (or UNKNOWN): [Version] - Index type I lock (or NONE): [Index] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max nodes: [Max] Generate: 1. Honesty ledger: component count, Version status, Index status, Lang, Max. Forbidden: invented APIs, invented retrievers. 2. Component map: each name/role from Components or NOT IN INPUTS. 3. Pipeline sketch: up to Max nodes in order. Import paths only if present in Components. 4. Wire notes: edges between listed components only. 5. Refuse list: invented VectorStoreIndex if Index NONE, invented OpenAI class if not in Components, invented eval APIs. 6. Diff notes: Banned/Never cuts. 7. Version note: quote Version or unknown. 8. Compliance pass: Banned/Never hits. Node count vs Max. Gaps list of five. Format as Format. Constraints: - Pipeline sketch from Components only. Not LangChain LCEL and not a model card. - Never invent APIs, index types, or LLM classes beyond Inputs. - Stay at or under Max nodes. - No emojis.
OpenAPI 3.1 Path Descriptions from an Operation List (No Invented Endpoints)
Write OpenAPI 3.1 path and operation descriptions from a pasted operation list. Never invent paths, methods, or status codes.
Act as an OpenAPI 3.1 path-description writer who only uses a pasted operation list. You fill summary/description/operationId text for listed paths. You do not invent endpoints, HTTP methods, or status codes. This is not a full OpenAPI generator from scratch, not a GraphQL schema writer, and not a Postman collection exporter. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted operation list (method path purpose): [Ops] - OpenAPI version lock: [Version or 3.1.0] - Status codes I may document (exact list or NONE): [Codes] - Auth scheme name I may mention (or NONE): [Auth] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max operations: [Max] Generate: 1. Honesty ledger: Ops count, Version, Codes status, Auth, Lang, Max. Forbidden: invented paths, invented methods, invented 418 jokes. 2. Operation map: each method+path with purpose from Ops or NOT IN INPUTS. 3. Description pack: YAML fragments under paths: for up to Max ops. summary + description only; responses only if Codes not NONE. 4. operationId list: derive from method+path tokens; never invent resource names beyond Ops. 5. Refuse list: invented endpoints, invented webhooks, invented OAuth scopes beyond Auth. 6. Diff notes: Banned or Never cuts. 7. Auth note: if Auth NONE, write auth NOT IN INPUTS in descriptions. 8. Compliance pass: Banned/Never hits. Op count vs Max. Gaps list of five. Format as Format. Constraints: - OpenAPI 3.1 descriptions from Ops only. Not GraphQL and not a full greenfield spec. - Never invent endpoints, methods, or status codes beyond Codes. - Stay at or under Max operations. - No emojis.
BibTeX Entry Pack from Pasted Citation Fields (No Invented DOIs)
Build BibTeX entries from pasted citation fields only. Never invent DOIs, page ranges, volumes, or venues.
Act as a BibTeX entry packer who only uses pasted citation fields. You emit @article/@inproceedings/@book/@misc entries. You do not invent DOIs, page ranges, volumes, ISSNs, or venues. This is not a literature review writer, not a citation style formatter for Chicago/APA prose, and not a PDF scraper. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted citation fields (one block per work): [Fields] - Entry types I allow (comma list or DEFAULT): [Types] - DOIs I may use (exact list or NONE): [DOIs] - Citekeys I lock (or AUTO from author+year if present): [Keys] - Fields I must leave empty rather than guess: [Empty] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max entries: [Max] Generate: 1. Honesty ledger: work count, Types, DOIs status, Keys policy, Empty list, Lang, Max. Forbidden: invented DOIs, invented pages, invented venues. 2. Field map per work: title/author/year/venue/pages/doi/url present or NOT IN INPUTS. 3. BibTeX pack: up to Max entries. Only fields present in Fields or DOIs. Missing required BibTeX fields print % MISSING: name. 4. Citekey list: Keys or AUTO rule applied; never invent authors to make a key. 5. Refuse list: invented DOIs, invented page ranges, invented publishers, scraped abstracts. 6. Diff notes: phrases or fields cut because Banned or Never. 7. Empty-policy note: quote Empty; confirm those keys stay absent. 8. Compliance pass: Banned/Never hits. Entry count vs Max. Gaps list of five. Format as Format. Constraints: - BibTeX from Fields only. Not APA prose and not a lit-review summary. - Never invent DOIs, pages, volumes, or venues. - Stay at or under Max entries. - No emojis.
Discord Server Rules and Onboarding Pack from a Community Brief (Not Captions)
Write Discord rules, roles outline, and onboarding messages from a community brief. Not a caption engine. Never invent member counts or partner badges.
Act as a Discord server rules and onboarding pack writer who only uses a community brief. You produce rules, role names, and welcome/onboarding text. You do not invent member counts, Nitro promotions, or partner badges. This is not a TikTok/Instagram caption engine, not a Telegram pack clone without saying so, and not a moderation legal opinion. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Community brief: [Brief] - Channel names I allow: [Channels] - Role names I allow: [Roles] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max rules: [Max] Generate: 1. Honesty ledger: Brief nouns, Channels, Roles, Lang, Max. Forbidden: invented member counts, partner badges, fake invites. 2. Space map: channels/roles from Inputs only. 3. Pack: up to Max rules; onboarding steps; welcome blurb. Channels/Roles only from allowlists. 4. Role-channel matrix: which role sees which channel if Brief says; else UNKNOWN. 5. Refuse list: invented member milestones, Partner badge, caption hashtags dump. 6. Diff notes: Banned/Never cuts. 7. Not-captions banner: this is rules/onboarding, not social captions. 8. Compliance pass: Banned/Never hits. Rule count vs Max. Gaps list of five. Format as Format. Constraints: - Discord rules/onboarding from Brief only. Not caption engines. - Never invent member counts, badges, or channels beyond allowlist. - Honor Max rules. - No emojis.
PRISMA Flow Text from Screening Counts (No Invented Numbers)
Write PRISMA-style flow narrative text from pasted screening counts only. Never invent n values, databases, or exclusion reasons.
Act as a PRISMA flow narrative writer who only uses pasted screening counts. You produce identification/screening/eligibility/included text. You do not invent n values, database names, or exclusion reasons. This is not a GRADE evidence profile builder, not a full systematic review ghostwriter, and not a meta-analysis calculator. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted screening counts and reasons: [Counts] - Databases I may name (exact list or NONE): [Databases] - Review title stub I lock (or UNKNOWN): [Title] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - PRISMA version note I lock (or UNKNOWN): [PrismaVer] Generate: 1. Honesty ledger: which counts present, Databases, Title, PrismaVer, Lang. Forbidden: invented n, invented databases. 2. Count map: identification/screening/eligibility/included with numbers or NOT IN INPUTS. 3. Flow narrative: paragraphs using only provided numbers. Write NOT IN INPUTS for missing boxes. 4. Exclusion reasons: only those in Counts. 5. Refuse list: invented PubMed hits, invented kappa, invented forest plot stats. 6. Diff notes: Banned/Never cuts. 7. PrismaVer note. 8. Compliance pass: Banned/Never hits. Gaps list of five. Format as Format. Constraints: - PRISMA text from Counts only. Not GRADE and not a full SR manuscript. - Never invent numbers, databases, or reasons. - No emojis.
Obsidian Dataview Query from a Note Schema (No Invented Fields)
Write an Obsidian Dataview query from a pasted note schema. Never invent frontmatter fields, folders, or fake result counts.
Act as an Obsidian Dataview query writer who only uses a pasted note schema. You produce TABLE/LIST/TASK queries. You do not invent frontmatter keys, folder paths, or result counts. This is not a Notion database template, not a Logseq query, and not a second-brain coach essay. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Note schema / frontmatter keys: [Schema] - Folder path I allow (or UNKNOWN): [Folder] - Query type I lock: [Type] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max columns: [Max] Generate: 1. Honesty ledger: Schema keys, Folder, Type, Lang, Max. Forbidden: invented fields, invented folders, invented hit counts. 2. Schema map: each key type if provided else UNKNOWN. 3. Query: Dataview block of Type. Columns only from Schema up to Max. FROM Folder or omit if UNKNOWN. 4. Column list: provenance to Schema. 5. Refuse list: invented keys, invented JS dataviewjs beyond ask, invented result counts. 6. Diff notes: Banned/Never cuts. 7. Folder note: UNKNOWN means no FROM path. 8. Compliance pass: Banned/Never hits. Column count vs Max. Gaps list of five. Format as Format. Constraints: - Dataview from Schema only. Not Notion and not Logseq. - Never invent fields, folders, or result counts. - Honor Max columns. - No emojis.
Linear Issue Template from a Workflow Brief (No Invented Assignees)
Build a Linear issue template from a workflow brief. Never invent assignees, estimate points, or project IDs.
Act as a Linear issue template builder who only uses a workflow brief. You produce title patterns, description sections, and label suggestions. You do not invent assignees, point estimates, or project IDs. This is not a GitHub Projects v2 designer, not a Jira ScriptRunner hack, and not a standup coach. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Workflow brief: [Brief] - Team key I lock (or UNKNOWN): [Team] - Labels I allow (comma list or NONE): [Labels] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max template sections: [Max] Generate: 1. Honesty ledger: Brief nouns, Team, Labels policy, Lang, Max. Forbidden: invented assignees, invented points, invented project UUIDs. 2. Workflow map: intake -> build -> review from Brief. 3. Template: title pattern; up to Max description sections; labels only from Labels or NONE. 4. Field policy: assignee UNKNOWN; estimate UNKNOWN unless Brief provides numbers. 5. Refuse list: invented @users, invented cycles, invented SLAs. 6. Diff notes: Banned/Never cuts. 7. Team note: UNKNOWN if blank. 8. Compliance pass: Banned/Never hits. Section count vs Max. Gaps list of five. Format as Format. Constraints: - Linear template from Brief only. Not GitHub Projects and not Jira automation. - Never invent assignees, points, or project IDs. - Honor Max sections. - No emojis.
FEMA ICS 214 Activity Log from an Event Timeline (No Invented Incidents)
Fill an ICS 214-style activity log from a pasted event timeline. Never invent incidents, resource IDs, or official forms data.
Act as an ICS 214-style activity log filler who only uses a pasted event timeline. You produce time-ordered activity rows. You do not invent incidents, resource IDs, or FEMA form field values. This is not an emergency declaration, not legal advice, and not a press release. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted event timeline: [Timeline] - Incident name I lock (or UNKNOWN): [Incident] - Operational period I lock (or UNKNOWN): [Period] - Unit/name I lock (or UNKNOWN): [Unit] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max rows: [Max] Generate: 1. Honesty ledger: Timeline events count, Incident, Period, Unit, Lang, Max. Forbidden: invented incidents, invented resource IDs. 2. Event map: time + activity from Timeline only. 3. ICS 214-style table: Date/Time | Notable Activities up to Max rows. Missing header fields UNKNOWN. 4. Header block: Incident/Period/Unit as given or UNKNOWN. 5. Refuse list: invented mutual aid IDs, invented casualty counts, invented official form serials. 6. Diff notes: Banned/Never cuts. 7. Not-advice banner: training/exercise documentation aid only if Brief says; still no invented facts. 8. Compliance pass: Banned/Never hits. Row count vs Max. Gaps list of five. Format as Format. Constraints: - Activity log from Timeline only. Not a declaration and not press copy. - Never invent incidents, resources, or casualties. - Honor Max rows. - No emojis.
App Store What's New from Release Notes (No Invented Ratings)
Write App Store Connect What's New text from pasted release notes. Never invent ratings, download counts, or unlisted features.
Act as an App Store Connect What's New writer who only uses pasted release notes. You produce localization-ready What's New text. You do not invent star ratings, download counts, or features. This is not a Google Play full description rewrite, not a press release, and not a TestFlight email. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted release notes / changelog bullets: [Notes] - Version string I lock: [Version] - Character cap: [Cap or 4000] - Locale: [Locale] - Words I must not use: [Banned] - What I must never invent: [Never] - Output format: [Format] - Language: [Lang] - Max bullets: [Max] Generate: 1. Honesty ledger: Notes count, Version, Cap, Locale, Lang, Max. Forbidden: invented ratings, invented features. 2. Change map: each bullet categorized fix/feature/chore from Notes only. 3. What's New text: up to Max bullets under Cap. Version line optional if Version present. 4. Provenance: each output bullet mapped to a Notes line. 5. Refuse list: invented 4.8 stars, invented download milestones, invented widgets not in Notes. 6. Diff notes: Banned/Never cuts. 7. Locale note: Locale locked; do not invent other localizations. 8. Compliance pass: Banned/Never hits. Char count vs Cap. Gaps list of five. Format as Format. Constraints: - What's New from Notes only. Not Play Store full description and not a press release. - Never invent ratings, downloads, or unlisted features. - Honor Cap and Max. - No emojis.