ShenyuanNext/wepr-growth-skills / skills/pr-strategy-workbench
pr-strategy-workbench
Verified规划和审校有证据支撑的公关策略、客户传播方案、媒体叙事、发布前风险控制、公开文案、新闻稿、声明、问答和声誉行动。适用于公关方案、传播策略、议题与舆情评估、竞品或媒体叙事分析、发布风险预案、发言人口径、公共语境文案审校和新闻稿撰写。正在发生的危机应转交 diagnose-pr-crisis。不得虚构事实、来源、引语、公众反应、法律结论或效果承诺。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 10.1 KB |
| agents/openai.yaml | config | 248 B |
| references/client-pr-plan.md | markdown | 2.7 KB |
| references/narrative-and-public-copy.md | markdown | 5.1 KB |
| references/research-prelaunch-and-annual.md | markdown | 5.3 KB |
references/research-prelaunch-and-annual.md
View on GitHub# Research, pre-launch, and annual PR planning
Use this reference only when the request requires consumer evidence conversion, industry media research, a pre-launch risk plan, or an annual PR strategy.
## Consumer evidence to PR action
Start with the decision the evidence must change. Separate verified facts, bounded sample signals, analyst explanations, and hypotheses. Do not turn qualitative frequency into prevalence, stated intention into purchase, a complaint into an incident rate, or a concurrent change into causality.
Translate only relevant evidence through five layers:
1. **Fact:** what people did or said, what the product delivered, and what the market evidence actually measures.
2. **Feeling:** expectation, anxiety, trust, unfairness, identity, loss, and respect, without substituting emotion for fact.
3. **Public narrative:** the simple interpretation that may form around the action, marked as a hypothesis until discussion evidence supports it.
4. **Business effect:** purchase, retention, price acceptance, channel support, product trust, complaint cost, or category understanding.
5. **Action:** act now, validate first, pause the unsupported direction, or take no external action.
For each proposed claim, label it as safe to say publicly, useful only for internal judgment, not currently supportable, or requiring decisive validation. Record the strongest counterevidence and what would reverse the recommendation.
## Industry media research
Define the research question, industry boundary, geography, language, time window, media/source types, depth, and completion condition. Build two ledgers before conclusions:
- outlet/source ledger: information role, audience, format, ownership or commercial disclosure when verifiable, coverage status, and inclusion reason;
- article sample ledger: author, title, date, original URL, original/repost/cooperation status when disclosed, topic, trigger, structure, sources, evidence processing, and access limits.
Use consistent coding for distribution, topics, writing choices, structure, evidence, and the industry understanding produced. Generate categories from repeated mechanisms and reader jobs; do not predetermine the number of categories or rank outlets by a composite score. Compare the same event across outlets, search for counterexamples, and separate article-, author-, section-, and outlet-level conclusions.
Suggested evidence-processing labels describe what a sampled item does:
- carries or lightly edits available material;
- adds context, terminology, or limited checks;
- adds independent interviews, data, or multi-source organization;
- performs multi-party investigation, field verification, and systematic challenge.
These labels are not quality grades. Report sample counts, window, retrieval date, unavailable content, and commercial-relationship blind spots with every pattern.
## Pre-launch contingency planning
First state the action's primary objective: attention, trust, category education, repair, authority, public legitimacy, or conversion support. Then scan six areas:
1. the action, pricing, promise, posture, and value position;
2. product, service, content, event, or delivery failure modes;
3. users, employees, partners, channels, and affected groups;
4. law, regulation, platform policy, and industry rules requiring qualified review;
5. brand, category, partner, or spokesperson history that may be reactivated;
6. supplier, creator, platform, competitor, and other risk transmission.
For each material item record evidence status, likelihood, impact, affected stakeholder, earliest signal, owner, prevention, response, approval, and trigger. Sort the outcome into:
- **must change the underlying action:** facts, harm, delivery, or organizational support fail;
- **must prepare a response:** action may proceed but predictable questions need evidence and accountable wording;
- **monitor:** low-materiality or weakly supported risk with explicit escalation conditions.
Deliver the likely challenge, a response the named organization could publicly stand behind, the required change/preparation/monitoring action, decision deadline, and go/no-go condition. Do not invent attacks or fill every scan category for appearance.
## Annual PR strategy
Classify the assignment as media relations, reputation/public-opinion management, or integrated PR, and as new drafting, revision, review, or section completion. Mark the document as framework, working, or decision-ready based on evidence and approvals.
Build one line of logic:
`business problem → annual purpose → priority stakeholders and risks → strategic narrative → programs and moments → phases → owners/resources → measurement → adaptation triggers`
Each workstream needs evidence, audience, intended observable change, deliverables, owner, approval, timing, dependency, KPI definition, trigger, and trust risk. Keep internal facts, external facts, analysis, forecasts, and missing inputs separate. Use explicit placeholders for missing evidence, organizational decisions, and untested assumptions; do not use public research to fabricate internal budgets, events, commitments, contacts, or baselines.
End with management decisions required, resource and approval constraints, scenario triggers, and a gap table explaining which missing input affects which judgment. Do not call a framework or working draft decision-ready.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |