Home/Blog/How to Draft a Semantic Kernel Plugin Skill Brief
Blog

How to Draft a Semantic Kernel Plugin Skill Brief

P
promptstudio
How to Draft a Semantic Kernel Plugin Skill Brief

Plugin briefs invent HTTP paths. Lock the tools list. Leave VERSION UNKNOWN when blank.

The matching generator is the Semantic Kernel Plugin Skill Brief from a Tools List prompt. Browse related cards in the PromptDig library (Browse more prompts). When a filled run survives, share the version you actually use (Share a prompt).

Lock Semantic Kernel version or mark UNKNOWN

Draft a Semantic Kernel plugin/skill brief from a tools list. No invented APIs, endpoints, or package versions. Start by filling Inputs, not by asking the model to remember last week's run. If a field is blank, write NONE or NOT IN INPUTS and leave it blank through Generate. The card is built so the model cannot honestly invent a number, owner, URL, or command that you did not paste.

Paste these fields before you hit run:

Tools list (name, purpose, params allowed): [Tools]
SK version I allow (or UNKNOWN): [Version]
Language/runtime I allow: [Runtime]
Plugin name I allow: [Plugin]
Words I must not use: [Banned]
What I must never invent: [Never]
Output format: [Format]
Language: [Lang]

That inventory is the honesty ledger. Anything that does not appear there is forbidden in the draft. If you catch yourself adding a nice-to-have after the run, you are no longer using the card. You are ghostwriting. Put the extra fact in Inputs and run again.

Table every skill from the tools list

Generate is numbered on purpose. Do not skip a step because the first paragraph looked done. The early steps exist to stop later prose from smuggling claims.

Walk the Generate list in order:

  1. Honesty ledger: Tools nouns, Version, Runtime, Plugin, Lang.
  2. Version lock: print Version or VERSION UNKNOWN.
  3. Plugin overview: Plugin name + one-liner purpose from Tools only.
  4. Skill table: one row per tool in Tools with name, description, params (only params listed).
  5. Missing params: for any tool without params, write PARAMS NOT IN TOOLS LIST.
  6. Auth/config: only if Tools names secrets/config keys; else AUTH/CONFIG NOT IN TOOLS LIST.
  7. Refuse: inventing HTTP paths, inventing NuGet/PyPI versions, inventing SK connectors not listed.
  8. Compliance pass: Banned and Never. Format as Format.

If a step asks for a version lock, quote the version from Inputs in the output. If a step asks for a refuse list, keep the refuse list in the published artifact, not in a sidebar you delete. Reviewers should see what the model was not allowed to do.

Skip auth when Tools omit secrets

Most failures are the same shape: a missing field gets a confident fill. A conversion rate appears. A Gradle task appears. A flash point appears. A caption appears on a job that asked for slide text only. Your review is to search the draft for numbers, names, and commands, then grep Inputs. No match means cut.

Honor the constraints as hard stops, not vibes:

  • Semantic Kernel brief from Tools list only. No invented APIs.
  • Never invent endpoints, SDK methods, or package versions.
  • Keep Version UNKNOWN when blank.
  • No emojis.

When the card says not legal advice, not certification, not an exam dump, or not a caption engine, that sentence belongs at the top of the output. Deleting it to look more finished is how you inherit risk.

Refuse invented endpoints and NuGet versions

Finish with the compliance pass the prompt already asks for. Quote the banned-word hits. Cut them. Print character counts when the job has a cap. Print word counts when the job has a budget. List gaps as gaps. Five missing facts are more useful than one smooth paragraph.

Tags on the card (semantic kernel plugin brief, skill brief from tools list, semantic kernel no invented apis) are a reminder of the job shape, not an invitation to wander into a neighboring cluster. If you need a different surface, open a different PromptDig card rather than stretching this one.

Fill the card, then run

Replace every bracket. Run on ChatGPT, Claude, or Gemini. Read the ledger first, then the artifact. If the model invents a commit, KPI, DOI, PEL, bid, or logo, discard the run. Tighten Inputs. Run again. Share the filled card that survived, not the first draft that sounded done.