Benchmark To Brief
Convert validated short-form video benchmark research into fact-grounded campaign briefs, hooks, and test plans without inventing unsupported claims.
Overview
Benchmark To Brief is an agent skill most often used in Validate (also Launch distribution planning) that turns validated benchmark research into fact-grounded short-form video campaign briefs and concept candidates.
Install
npx skills add https://github.com/postplusai/postplus-skills --skill benchmark-to-briefWhat is this skill?
- Hard fact rule: every angle, persona, and claim must trace to supplied research artifacts
- Structured output JSON: corePromise, hookOptions, workflow, sourceFacts, sourceBasis
- Produces briefs, concept lists, hook libraries, lane priorities, and test matrices—not final scripts
- Explicit handoff: script-ready brief to human script review; visuals only after evidence-backed brief
- PostPlus Media and Creative Production family skill for short-form video campaigns
- Structured artifact fields: corePromise, hookOptions, workflow, sourceFacts, sourceBasis
Adoption & trust: 1 installs on skills.sh; 8 GitHub stars; 3/3 security scanners passed (skills.sh audits); trending (+100% hot-view momentum).
What problem does it solve?
You have research tables and reports but no disciplined brief that ties hooks and concepts to real evidence before spending on scripts or visuals.
Who is it for?
Builders running evidence-led short-form campaigns who already invested in benchmark research and need a strict research-to-strategy handoff.
Skip if: Greenfield brainstorming without sources, final script writing, or direct visual generation before a brief exists.
When should I use this skill?
You already have benchmark research artifacts and need campaign briefs, concepts, hooks, or test plans grounded in that data.
What do I get? / Deliverables
You get a structured, source-cited brief and concept set ready for human script review or downstream asset generation—without fabricated angles.
- Evidence-backed campaign brief
- Concept candidate and hook option lists
- Lane priorities and test matrices for short-form video
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Validate because the skill assumes research artifacts already exist and turns evidence into scoped creative strategy before production. Scope fits tightening research into briefs, lanes, and test matrices rather than shooting or editing footage.
Where it fits
Narrow ten researched patterns into three testable concept lanes with cited hook options.
Prioritize distribution tests using a brief matrix grounded in performance findings.
Refresh hook libraries when new comment analyses arrive without drifting from source facts.
How it compares
Evidence-locked brief compiler—not a generic creative brainstorm or render skill.
Common Questions / FAQ
Who is benchmark-to-brief for?
Solo creators and indie teams using PostPlus-style agent workflows who validate TikTok/Reels/Shorts patterns before production.
When should I use benchmark-to-brief?
After benchmark research is done in Validate to scope campaigns; before Launch distribution tests when you need hook libraries and test matrices tied to data.
Is benchmark-to-brief safe to install?
It processes your research artifacts locally in the agent context; review the Security Audits panel on this Prism page for the upstream package.
SKILL.md
READMESKILL.md - Benchmark To Brief
# Benchmark To Brief ## Use When - Turn validated benchmark research into campaign briefs, concept candidates, hook libraries, lane priorities, or test matrices. - Use after research already exists; this is the strategy handoff between evidence and execution. ## Do Not Use When - The task is greenfield brainstorming without source evidence. - The user needs final scripts; return a script-ready brief and send it to the human script review gate. - The user needs visual generation; hand off to asset generation workflows after the evidence-backed brief exists. ## Required Input - Validated evidence such as benchmark reports, master tables, pattern tables, strategy tables, comment analyses, or performance findings. - For the structured artifact, input JSON with `corePromise`, `hookOptions`, `workflow`, `sourceFacts`, and `sourceBasis`. ## Fact Rule Everything produced by this skill must be grounded in available evidence. Do not invent: - target personas with no source basis - hooks that contradict benchmark wording - unsupported product claims - audience motivations not reflected in data If evidence is incomplete, separate: - `Observed from sources` - `Inference` - `Open question` ## Default Workflow 1. Use the smallest sufficient evidence set already supplied or explicitly identified by the user. 2. Prefer structured artifacts first: final report, strategy table, pattern table, master table, then comment summaries. 3. Extract facts before proposing: winning lanes, repeated hook shells, repeated structure types, visual format tendencies, user-language patterns, and strong examples. 4. Map facts to brief fields: source artifact, benchmark ids or examples, why the pattern fits the product, and variable being tested. 5. Keep each concept narrow: one problem, one hook family, one format, one test variable. 6. Preserve benchmark language when adapting hooks unless the user asks for variant testing. 7. Return an execution object: brief, concept list, hook set, script input, or approved copy. ## Judgment Flow For every recommendation, answer: - what problem is being cut? - what hook shell is being reused? - what content lane does it belong to? - what variable is being tested? - which source evidence supports it? - what claim or audience motivation is only inference? ## Stop Conditions - Stop when required user intent, source evidence, or owned input artifacts are missing and guessing would change the result. - If an owned CLI or script command fails, report the exact error and stop. Do not bypass the failure with metadata-only answers, readiness probing, local payload rewrites, fallback providers, or unpublished tools. ## Output Shape Common outputs: - campaign brief - concept candidates - hook library - lane prioritization - AB test matrix Campaign brief fields: - `objective` - `priorityLanes` - `personaConstraints` - `approvedHookFamilies` - `prohibitedDirections` - `variablesToTest` - `sourceBasis` Concept candidate fields: - `conceptId` - `problemToCut` - `hookType` - `hookDraft` - `targetLane` - `formatType` - `whyThisFits` - `sourceBasis` - `testVariable` ## Anti-Patterns - vague ideas like "make a productivity video" - copying benchmark content without adapting the concrete product variable - rewriting concrete benchmark language into abstract brand claims - treating raw research as execution-ready without a brief