Generate original British political satire sketches in the pompous-establishment two-hander tradition, a formal interview, briefing, consultation scene built on faux-serious rhetoric, class pretension, institutional hypocrisy, euphemism, slow-burn escalation, self-serving reasoning concealed beneath polished language. Use this skill for an original satirical sketch script, an interview-format dialogue exchange, a briefing-scene transcript, featuring a pompous official, minister, executive, regulator, spokesperson, peer, who sincerely believes their evasive reasoning is sound. Not intended for reproducing real sketches, quoting real public figures' actual recorded words, near-verbatim programme dialogue, slapstick, absurdist non-political comedy, or monologue writing without an interview/dialogue frame.
71
86%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
The comedy works because sincere, articulate self-justification by a powerful character is funnier and more damning than any external mockery — the character must never realise they have said anything absurd, or the satire collapses into simple insult.
Only use this skill when the request calls for an original two-hander, interview, briefing, or consultation-style satirical scene — not for a single joke, a monologue, or real-person mimicry.
Work through these steps in order when drafting a new sketch; you MUST complete step 8 before treating any draft as finished.
Beyond the Workflow steps above: satire targets are class, institutional privilege, political hypocrisy, official language, and self-serving explanations — you don't need to hit every one in a single piece. Contrasting refined language with increasingly ridiculous implications is optional but effective, and the interviewer can either ask innocent-seeming questions or restate the character's own words back at them; either approach works.
The humour should be sophisticated, dry, political, and character-driven rather than random. Create original characters and material using these structural techniques — never named real individuals' actual dialogue.
PREFER sustained escalation over rapid-fire jokes, but AVOID treating that as a rigid rule: UNLESS the brief specifically calls for a single quick exchange, three to five turns is TYPICALLY enough, and the RECOMMENDED default is to let a promising exchange run rather than cutting it short.
NEVER reproduce actual Bird & Fortune sketch content, character names, or dialogue verbatim or near-verbatim.
SYMPTOM: a line of dialogue is a close paraphrase or verbatim quote of an existing sketch, or reuses a recognisable character name or premise from an existing programme.
WHY: the source material is copyrighted, and reproducing it defeats the purpose of a generative skill — the user receives a copy of existing work, not new satire built from the technique.
CONSEQUENCE: the output is unusable (infringing, and not what was asked for) and must be discarded and rewritten from scratch.
❌ BAD
[Reproducing a recognisable existing sketch's premise and character names nearly word-for-word]
✅ GOOD
INTERVIEWER: Minister, the report says the scheme cost taxpayers eleven million pounds
to save four hundred thousand.
MINISTER: That's a very narrow way of expressing what was, in every meaningful sense,
a resounding demonstration of fiscal discipline.This shows an original premise (a fictional minister, an invented scheme) using the same technique — evasive arithmetic — without borrowing any existing character or line.
NEVER give the establishment character genuine self-awareness of their own hypocrisy.
SYMPTOM: the character explicitly acknowledges, even in an aside, that their own reasoning is dishonest, self-serving, or ridiculous.
WHY: the comedy depends on sincere, confident self-justification; a character who "gets the joke" collapses the satire into simple mockery, which is a weaker and less specific form of comedy.
CONSEQUENCE: the scene stops being satire and becomes a character simply confessing wrongdoing, which reads as flat rather than funny.
❌ BAD
EXECUTIVE: Look, between us, I know this makes me sound awful, but the shareholders
come first, don't tell anyone I said that.
✅ GOOD
EXECUTIVE: The shareholders' interests and the public interest are, properly understood,
the same interest — one simply requires a longer lens than the other.The GOOD version keeps the executive sincerely convinced they've said something wise, which is where the laugh lives.
NEVER use crude or direct insults in place of euphemism.
SYMPTOM: a character's line states an unflattering fact bluntly (an insult, a plain confession) rather than concealing it in polished or bureaucratic language.
WHY: the comic tension comes from polished, evasive language concealing something obviously indefensible; blunt language removes that tension and turns wit into abuse.
CONSEQUENCE: the piece loses its "establishment" register and reads as generic mockery rather than sustained satirical irony.
❌ BAD
OFFICIAL: We sacked her because she's an idiot and we don't like her.
✅ GOOD
OFFICIAL: Her departure reflects a mutual recognition that her talents were, at this
juncture, more suited to opportunities elsewhere.The GOOD version preserves the euphemistic concealment that the rest of the technique depends on.
[Faux-serious interview frame]
INTERVIEWER: So to be clear, the consultation lasted six minutes.
REGULATOR: Six substantive minutes, yes, following eighteen months of preparatory
listening.This demonstrates the frame technique: an ordinary interview question exposes a gap the regulator immediately tries to paper over with a technical distinction.
Expected: the laugh lands on the gap between "six minutes" and "eighteen months of preparatory listening," not on any explicit insult.
[Class pretension]
PEER: One does try to keep an open mind about state education, in the abstract.The comic mechanism here is the character's confident, distancing vocabulary ("one", "in the abstract") signalling unexamined privilege without any direct statement of prejudice.
[Euphemism concealing absurdity]
SPOKESPERSON: The figures weren't wrong, exactly — they were forward-looking in a way
that hadn't yet been achieved by events."Forward-looking" stands in for "false", showing how bureaucratic phrasing can make an admission of error sound like a virtue.
[Slow-burn escalation across three exchanges]
INTERVIEWER: Did the department know about the shortfall in March?
OFFICIAL: We had an early awareness.
INTERVIEWER: How early?
OFFICIAL: Early enough to be described, in due course, as timely.
INTERVIEWER: So you knew in March.
OFFICIAL: I wouldn't want to commit to a single month owning the whole truth.Each answer concedes slightly more while the official's confidence never drops — the escalation is in the audience's growing certainty, not in the character's tone.
[Calmly stated absurdity as an ending]
MINISTER: In hindsight, closing the only bridge before building the second one was,
I'll concede, a sequencing matter we are still learning from.Ending on a mild, procedural word ("sequencing matter") for what is an obvious catastrophic error is the "calmly stated absurdity" landing described in the workflow.
If a drafted sketch isn't landing, diagnose by symptom rather than rewriting from scratch:
references/euphemism-and-escalation.md and commit to it throughout.references/euphemism-and-escalation.md.Run a final check against these items before sharing the sketch; if any check fails, return to the matching Troubleshooting entry above:
| Topic | Reference | When to Use |
|---|---|---|
| Euphemism and escalation techniques | Euphemism and Escalation | When the concealment language feels too obvious, too crude, or the escalation is happening too fast across only one or two lines |
| Interview frame and persona construction | Interview Frame and Personas | When starting a new sketch and you need to fix the character's institutional role, register, and interviewer dynamic before writing any dialogue |
a1083f4
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.