Public Relations
Corey HainesCommunityCreated by Corey Haines
- Skill
- Workflow
- Business and productivity
- MIT
- No issues found
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
This skill provides a structured approach to public relations for software products, focusing on earned media as a multiplier for distribution rather than a primary conversion driver. It covers the full lifecycle of PR, from proactive pitching and newsjacking to inbound media requests and crisis management. The skill emphasizes building authentic relationships with journalists, crafting data-driven stories, and maintaining a professional media presence through owned assets like press pages and media kits.
What it does
- Guides users through proactive journalist pitching and media list building.
- Provides frameworks for reactive PR, including newsjacking and responding to media inquiries.
- Offers templates and playbooks for crisis communications and incident response.
- Defines best practices for creating and maintaining a company press page and media kit.
- Outlines strategies for podcast guest preparation and thought leadership placement.
- Establishes a quality bar for pitches to ensure relevance and effectiveness.
When to use it
- When you have a compelling story, such as proprietary data, a unique opinion, or a significant milestone.
- When you need to respond to a PR crisis, data breach, or negative public feedback.
- When you want to build long-term media relationships and brand authority.
- When you are preparing for a podcast appearance or an op-ed submission.
- When you need to set up foundational PR assets like a press page or media kit.
Advantages
- Focuses on high-impact, sustainable PR strategies rather than vanity metrics.
- Provides actionable templates for various PR scenarios, reducing drafting time.
- Includes specific guidance for crisis management to minimize reputational damage.
- Emphasizes the importance of journalist-centric pitching to increase response rates.
- Integrates PR into the broader marketing and distribution strategy.
Usage examples
- “I have a new data set about AI adoption in finance. How do I pitch this to a journalist?”
- The agent will guide you to the data story template, help you identify relevant journalists who cover fintech, and advise on how to structure the pitch to highlight the most interesting patterns.
- “We just had a minor service outage and users are complaining on social media. What should I do?”
- The agent will help you assess the severity tier, draft a holding statement, and provide a checklist for the first 60 minutes of crisis response.
- “How do I build a media list for my SaaS product?”
- The agent will provide a discovery checklist, suggest methods for finding relevant journalists, and explain how to score them based on beat match and reach.
- “I'm going to be a guest on a popular tech podcast next week. How should I prepare?”
- The agent will walk you through researching the host and show, mapping your core stories to their recurring themes, and creating a brief to ensure your appearance is impactful.
Install
Download the skill as a ZIP:
https://autogineer.com/agent-skills/coreyhaines31/public-relations/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip public-relations.zip -d .agents/skills
Documentation by Corey Haines
Public Relations & Earned Media
You are an expert in earned media for software products. Your goal is to help the user get covered by journalists, podcasts, and newsletters — efficiently, with respect for the people on the other end of the pitch.
Before Starting
Check for product marketing context first:
If .agents/product-marketing.md exists (or .claude/product-marketing.md, or the legacy product-marketing-context.md filename, in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
Core Philosophy
PR is not a substitute for distribution. It's a multiplier for it.
Earned media doesn't drive direct conversions. A TechCrunch hit will not give you 1,000 paying customers. It will give you backlinks, brand legitimacy, AI-citation surface area, and ammo for sales conversations.
Pitch journalists like you'd pitch a customer: specific, useful, fast, and never about you.
The story is not your product. The story is the trend, the data, the conflict, or the human. Your product is the evidence. Every pitchable story bends toward one of three angles — Founding Story, David vs Goliath, or Have an Enemy (a broken system, never a competitor). See references/story-angles.md.
Chase press for the compound effect, not the traffic bump. The bump fades in a day; authority, journalist relationships, and AI-citation surface compound. Build media relationships before you need them, and run one core asset through the whole repurposing flywheel.
Speed beats polish on reactive PR. A B+ pitch in the first hour of a story beats an A+ pitch on day three.
When PR is worth it
You have a real story — proprietary data, a strong opinion, a milestone, a customer with a sharp before/after, or a fresh angle on a trending topic
You have founder/exec time — journalists want quotes from people with skin in the game, not from a PR rep
You have a destination — a press page, blog post, or product launch that converts attention into something useful
When to skip PR (for now)
Pre-launch with no story beyond "we exist"
No one on the team can sustain pitching for 4–6 weeks (PR is a momentum game)
You don't have a clear ICP — journalists ask "who reads my piece because of this?" and if you can't answer, neither can they
The PR Mix
Four modes. Most teams over-index on one. Run at least three.
| Mode | What it is | Effort | Speed to coverage |
|---|---|---|---|
| Reactive (newsjacking) | Inject your POV into trending news | Low–medium | Hours to days |
| Proactive (pitching) | Build a media list, pitch original stories | High | 2–8 weeks |
| Inbound (press requests) | Respond to journalist queries on HARO/Qwoted/Featured | Low | Days to weeks |
| Owned (press page + media kit) | Make it easy for journalists to find you | One-time setup | N/A |
For the story angle taxonomy (Founding Story / David vs Goliath / Have an Enemy), data stories, media relationship-building, and the PR repurposing flywheel — see references/story-angles.md
For the reactive newsjacking workflow — see references/newsjacking.md
For proactive journalist pitching — see references/journalist-pitching.md
For inbound press-request platforms (HARO, Qwoted, etc.) — see references/press-platforms.md
For where to pitch (media outlets, podcasts, newsletters) — see references/media-outlets.md. For startup/SaaS/AI directories, use the separate directory-submissions skill — different intent, different list.
For prepping a podcast appearance you've landed — see references/podcast-guest-prep.md. Episodes get transcribed and cited by AI assistants, so a good appearance compounds in AI answers for years — prep is an AI-visibility play, not just interview polish.
When Something Goes Wrong
PR also has a reactive side: breaches, outages, viral complaints, executive controversies. The story is already running, so the goal shifts from earning coverage to responding fast and accurately.
Start by assigning a severity tier, then run the first-60-minutes checklist: confirm facts, designate one spokesperson, publish a holding statement, set the next update time. For a data breach, injury, lawsuit, or regulator, loop in legal counsel before admitting fault or publishing.
For severity tiers, channel order, and playbooks by crisis type — see references/crisis-communications.md
For holding, breach, outage, apology, billing-error, and postmortem templates — see references/statement-templates.md
Owned: Press Page + Media Kit
Set this up once. It's the cheapest PR investment with the highest ROI on every future story.
Press page (/press or /newsroom) should include:
One-paragraph company description (copy/paste ready)
Founder bios with headshots (high-res, downloadable)
Logo pack (SVG + PNG, light + dark, with usage guidelines)
Product screenshots (high-res)
Recent coverage list (social proof for the next journalist)
Founding date, employee count, funding (if disclosed)
Press contact email (not a form — journalists hate forms)
Recent press releases / announcements
One sentence at the top: "For interview requests or assets, email [email protected] — we respond within 24 hours."
Then actually respond within 24 hours.
Quick Reference: Pitch Quality Bar
Before sending any pitch, the answer to all of these should be yes:
[ ] Does this journalist cover this beat? (Check their last 5 articles.)
[ ] Is there a clear news hook — something that just happened or is about to?
[ ] Could this journalist write a complete story from this email alone? (Data, quotes, customer name, contact.)
[ ] Is the subject line specific enough to predict the article's headline?
[ ] Is the pitch under 150 words?
[ ] Did you avoid the words "revolutionary," "game-changing," "disruptive," and "synergy"?
[ ] Is the ask clear? (Interview? Embargo? Exclusive? Quote?)
If any answer is no, don't send.
Measurement
What to track:
| Metric | Why |
|---|---|
| Coverage count (placements / month) | Activity baseline |
| Domain rating of placements | Backlink value |
| Referral traffic from coverage | Did anyone actually click? |
| Brand search lift | Did people search you after reading? |
| AI citation rate (ChatGPT, Perplexity quote your brand?) | The new measurement that matters |
| Sales conversations citing the article | The only one that matters for revenue |
What not to obsess over: AVE (advertising value equivalency) — it's a vanity metric PR firms invented.
Common Workflows
"Help me newsjack [trending story]"
Go to newsjacking.md, run the scoring rubric, draft 2–3 angles, pick the best, draft the pitch.
"Find journalists who cover [beat]"
Go to journalist-pitching.md, use the discovery checklist + dev-browser to research recent articles, build a scored list.
"What's worth pitching this week?"
Combine: recent product milestones + active news cycles + any data you've collected. Score each potential story by the quality bar above.
"What's my story angle?" / "How do I get press with no news?"
Go to story-angles.md. Fit the situation to one of the three angles (Founding Story / David vs Goliath / Have an Enemy), or turn proprietary data into a data story. Remember: a milestone alone isn't a story — milestone with narrative is.
"Respond to this HARO query"
Go to press-platforms.md, use the response template, keep it under 200 words.
"I'm going on [podcast] next week — help me prep"
Go to podcast-guest-prep.md: research the show (RSS feed → site → Apple Podcasts → web), extract the recurring threads and host profiles, map the guest's stories onto them, deliver the brief.
"Build my press page"
Use the checklist above. Most companies do this in an afternoon and forget about it for a year — that's fine.
"Get us more backlinks"
PR earns the highest-ceiling links but takes the most effort. For reclamation, partner pages, best-X lists, and the rest of a link program, use the link-building skill and treat PR as one of its tactics.
Files
- ATTRIBUTION.mdAdded by AutoGineer893 B
- LICENSELicence file added by AutoGineer1.0 KiB
- SKILL.md9.6 KiB
- evals/evals.json3.7 KiB
- references/crisis-communications.md8.4 KiB
- references/journalist-pitching.md10.8 KiB
- references/media-outlets.md9.1 KiB
- references/newsjacking.md8.4 KiB
- references/podcast-guest-prep.md5.5 KiB
- references/press-platforms.md6.7 KiB
- references/statement-templates.md5.5 KiB
- references/story-angles.md6.4 KiB
references/crisis-communications.md
Crisis Communications
Adapted from a contribution by @fyscleaning-jpg (PR #381).
Reactive PR for when something has gone wrong: a breach, an outage, a viral complaint, an executive controversy. Proactive PR runs on your timeline. In a crisis, the story is already running, and your job is to respond quickly, honestly, and without making it worse.
Contents
Get the facts first
Legal, regulatory, and safety exposure
Principles
Severity tiers
The first 60 minutes
Channel order
Playbooks by crisis type
What not to do
After the crisis
Review checklist
Get the Facts First
Before drafting anything, establish:
What actually happened. The confirmed fact, separate from the rumor. If it's unconfirmed, the statement says so.
Who is affected. Customers, employees, the public, a specific group, regulators.
The exposure. Contained, still unfolding, or likely to get worse (more data exposed, more people harmed, outage ongoing).
What's already public. Breaking on social, covered by press, or known only internally.
What legal, security, or engineering can confirm right now versus what's still under investigation.
Never draft a statement that asserts something not yet confirmed. A statement you walk back 24 hours later does more damage than a slower, accurate one.
Legal, Regulatory, and Safety Exposure
If the incident involves a data breach, physical injury or safety risk, a lawsuit, or a regulator, loop in legal counsel before you admit fault, describe cause, or publish anything beyond a holding statement.
Data breaches can carry mandatory notification deadlines and required content, and these vary by jurisdiction and by the type of data involved. Counsel determines which apply. Don't state a specific deadline unless counsel has confirmed it.
Keep the legal posture and the public statement as separate documents. Counsel decides what you can say. The public statement says it in plain human language.
Principles
Speed matters, accuracy matters more. Silence in the first hours reads as incompetence or concealment. A wrong statement reads as dishonesty.
One source of truth. Designate a single spokesperson and a single channel of record. Mixed messages from different staff become their own crisis.
Acknowledge before you explain. Open with what happened and who it affects. Readers scan for "am I affected and what do I do."
Apologize without a "but." "We're sorry this happened, but our systems generally..." is a defense with an apology in front. Once counsel has cleared accepting responsibility, say sorry, say what you're doing, and stop.
Keep updating. A crisis is a sequence of updates. Commit to a cadence ("next update by 6pm ET") and keep it, even when the update is "still investigating."
Severity Tiers
Match the response to the actual severity. When unsure, tier up: under-responding to a Tier 3 looks worse than over-responding to a Tier 2.
| Tier | Examples | Response |
|---|---|---|
| 1. Contained | Single feature bug, isolated complaint thread, minor copy mistake | Support or community team replies directly. No public statement. |
| 2. Visible | Service degradation, a complaint gaining traction, a marketing claim called out | Public acknowledgment on status page or social, owned by the support or marketing lead |
| 3. Significant | Full outage, billing error across many customers, a widely shared customer-harm story | Formal exec-reviewed statement on status page, email, and social. Spokesperson designated. |
| 4. Severe | Data breach, security incident, executive misconduct, safety issue, regulatory inquiry | Counsel-reviewed statement, exec-level spokesperson, dedicated incident page. Notification obligations may apply. |
The First 60 Minutes
Confirm the facts with the people who have direct knowledge (engineering, security, legal, support lead), not secondhand summaries.
Assess the tier. For Tier 4, bring in legal counsel now.
Designate the spokesperson and the channel of record (status page, pinned post, or both).
Publish a holding statement. It says you know about the issue and are working on it, with zero unconfirmed claims about cause or scope. See statement-templates.md.
Notify staff before or alongside going public. A short internal note covers what's happening, what to say if asked, and where to route press and customer questions.
Set the next update time and state it publicly.
Channel Order
Getting the order wrong, such as posting publicly before telling affected customers, creates a second crisis.
Directly affected parties first. Individual email or in-app notice to the people actually impacted, especially for breaches and billing errors.
Internal note, in parallel with step 1.
Status page or incident page. The linkable source of truth you control and update.
Owned social channels. A short post linking to the full statement. Character limits force omissions that read as evasive.
Press. If reporters are already asking, respond directly instead of waiting. Press should never get the news before affected customers.
Playbooks by Crisis Type
| Type | First move | Watch for |
|---|---|---|
| Data breach / security incident | Confirm scope with security and bring in counsel before any public claim about what was or wasn't exposed. Counsel checks notification obligations. | "No sensitive data was affected" is the statement most often walked back. Don't say it until security confirms it. |
| Product outage | Status page update within minutes, even if it only says "investigating." | Don't commit to an ETA you can't keep. "Next update by [time]" is safer than a guessed fix time. |
| Executive misconduct | Legal and board involvement before any statement. Here, getting it right outranks speed. | A statement that reads as protecting the executive over the people affected. |
| Customer-harm story going viral | Contact the customer privately before replying publicly. | Arguing facts in public reply threads. Resolve privately, then post a brief update if appropriate. |
| Billing / pricing error | Confirm scope (how many customers, how much money) before announcing the fix. | Waiting to be asked. Refund or credit proactively; refunds after public pressure read as conceding under fire. |
| Layoffs leak early | Move the internal announcement up so employees hear it from the company. | Letting the leak stand without an official statement. |
| Misleading marketing claim called out | Name the specific claim, correct it, and explain the fix. | Defending intent. The audience cares about the correction. |
What Not to Do
Go silent. "No new information yet, next update at [time]" beats nothing.
Delete criticism. Deleted threads get screenshotted and become a second story about a cover-up.
Issue a non-apology. "We're sorry if anyone was affected" puts the burden on the reader's reaction instead of owning the action.
Speculate. "We believe this may have also affected..." starts a correction cycle. State what's confirmed and name what's still under investigation.
Admit fault or cause before the facts and counsel allow it. You can acknowledge the issue and the impact without assigning blame you can't yet support.
Let legal language into the human statement. "The Company denies any wrongdoing" inside a customer apology reads as insincere.
Bury the lede. Don't open with paragraphs of context before saying what happened and who's affected.
After the Crisis
Public postmortem for Tier 3 and 4: what happened, root cause, and what's changing. Specific, checkable commitments build trust; "we take this seriously" doesn't.
Internal retro, separate from the postmortem, on what should change in the response process itself.
Track every promise. If the statement promised a fix, credit, or policy change, see it through and post a follow-up when it ships.
Watch sentiment for 1-2 weeks. Crises often get a second wave when a new detail surfaces.
Review Checklist
Does the response match the tier?
Have directly affected parties been notified before any public statement?
Does the draft claim only what's confirmed?
For breaches, injuries, lawsuits, or regulators: has counsel reviewed it?
Is there one spokesperson and one channel of record?
What's the next update commitment, and who owns it?
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Security
No issues found by our static scan· No issues found
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Created by
- Corey Haines
- Publisher
- Corey Haines (Community)
- Source
- coreyhaines31/marketingskills · skills/public-relations · commit 96798cf
- Licence
- MIT License (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Corey Haines or Corey Haines.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download public-relations.zip