@talsraviv/comedy-writers-room
BUse when helping write stand-up comedy, jokes, or humorous material about any topic
Install
agr install @talsraviv/comedy-writers-room --target claudeWrites 6 files into .claude/skills/, pinned to git-b14434f9.
- .claude/skills/comedy-writers-room/README.md
- .claude/skills/comedy-writers-room/SKILL.md
- .claude/skills/comedy-writers-room/audience-enthusiast-prompt.md
- .claude/skills/comedy-writers-room/audience-overthinker-prompt.md
- .claude/skills/comedy-writers-room/audience-skeptic-prompt.md
- .claude/skills/comedy-writers-room/writer-prompt.md
Document
name: comedy-writers-room description: Use when helping write stand-up comedy, jokes, or humorous material about any topic
Comedy Writers Room
Write jokes using subagents as a simulated audience. One agent writes, three agents react as different personas, then iterate based on their feedback.
Process
Step 1: Dispatch the writer
Read writer-prompt.md and use that template to dispatch a subagent.
- Replace
[TOPIC]with the user's topic - For first draft, omit the revision section
Step 2: Dispatch all 3 audience members in series
Read each audience template and dispatch them one at a time:
audience-enthusiast-prompt.mdaudience-skeptic-prompt.mdaudience-overthinker-prompt.md
Replace [JOKES] with the jokes from the writer.
Step 3: Synthesize feedback
Read what each persona said. What landed? What fell flat? What confused people?
Step 4: Dispatch writer again with feedback
Use writer-prompt.md again, this time:
- Include the revision section
- Replace
[FEEDBACK]with the synthesized audience reactions
Step 5: Repeat steps 2-4 until the jokes land
Stop when the personas are reacting well.
TESTING: Limit to 1 round of feedback for now.
Step 6: Present final material to user
Include a brief summary of how it evolved - what got cut, what emerged from iteration.
Critical Rules
- Always use Task tool - Never shell out to
claudeCLI - Dispatch audience in series - One at a time, so each reaction is visible
- Include feedback in revisions - The writer needs to know what to fix
Trustgrade B
- passBody integrity
Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.
- passType matchnot applicable to this artifact type
Whether the artifact is really the kind of thing its metadata claims it is.
- passFreshness
How long since the source repository was last pushed to.
- passPrompt injection
Scans the artifact's own text for instructions aimed at your agent rather than at you.
- warnLicenseno SPDX license detected
Whether the source repository declares an SPDX license permissive enough to redistribute.
How the grade is calculated
Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:
- Aevery check passed
- Bone warning
- Ctwo warnings
- Dprompt injection or body integrity failed, or three warnings
- Fone of those failed, and something else is wrong
These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.
Versions
git-b14434f9e28d2026-07-31