JasonColapietro/suede-creator-skills / skills/suede-public-relations
suede-public-relations
VerifiedBuild earned-media angles, sourced media lists, journalist and podcast pitches, reporter replies, and press kits with consistent entity facts. Use when asked to "draft a journalist pitch", assess newsjacking, or prepare third-party coverage for search and AI citations. Returns review-ready drafts and measurement steps. NOT FOR: directory submissions (use suede-directory-submissions), release packaging (use suede-launch-packaging), social posts (use suede-social), or sending without approval.
| File | Kind | Size |
|---|---|---|
| CARD.md | markdown | 4.3 KB |
| SKILL.md | skill md | 11.8 KB |
| agents/openai.yaml | config | 590 B |
| evals/evals.json | config | 2.8 KB |
| references/journalist-pitching.md | markdown | 12.6 KB |
| references/media-outlets.md | markdown | 6.3 KB |
| references/newsjacking.md | markdown | 9.3 KB |
| references/press-platforms.md | markdown | 7.3 KB |
| references/worked-handoff.md | markdown | 1.6 KB |
references/journalist-pitching.md
View on GitHub# Journalist Pitching: Proactive PR Workflow
Building a media list, scoring journalist fit, and crafting draft pitches for
review. Set the test duration from current story flow, team capacity, and
measured response rather than a universal campaign window.
## Contents
- Building the media list
- Scoring journalist fit
- Pitch templates by angle
- Subject lines that get opened
- Voice and structure
- Embargoes, exclusives, and follow-up etiquette
- Pitch killers
- Tooling
---
## Building the Media List
Build the smallest source-backed list that covers the relevant beat and current
story. Set its size from recent coverage volume and review capacity, not a fixed
quota or a database export.
### Discovery checklist
For each candidate journalist:
- [ ] Review a bounded recent sample from the author page; stop when the sample
is enough to establish or reject current beat fit
- [ ] Note their **publication**: does it reach your ICP?
- [ ] Check their **bio** on the outlet site: what topics do they own?
- [ ] Check **X/LinkedIn** for what they're posting about this week
- [ ] Note their **email** (usually on outlet author page, Muck Rack, or company About page)
- [ ] Check **Muck Rack** if available: it shows recent topics and pitch preferences
### Where to find candidates
| Method | How |
|--------|-----|
| **Reverse lookup from coverage you want** | Review a bounded recent article sample and note relevant bylines |
| **Topic search on Muck Rack** | Free tier shows journalists by topic |
| **X / Twitter lists** | "[your niche] reporters" lists already exist |
| **LinkedIn search** | "Journalist" + "[your category]": filter by recent activity |
| **Newsletter author pages** | Beehiiv, Substack, ConvertKit creators are pitchable |
| **Podcast host research** | Review a recent episode or transcript when it is needed to verify topic and format fit |
### Don't waste time on
- **Mass media databases** (Cision, Meltwater) for early-stage: overkill and expensive
- **Journalists who haven't posted in 6+ months**: they may have left
- **"Editor-in-chief" generic addresses**: pitches there get ignored or routed to interns
- **Journalists who explicitly state "no PR pitches" in bio**: respect it
---
## Scoring Journalist Fit
Score each journalist 1–10 across four dimensions, sum, and rank. Retain only
the candidates supported by current evidence and the approved review capacity.
| Dimension | What it measures | Weight |
|-----------|------------------|--------|
| **Beat match** | Do they cover your category specifically? | 3x |
| **Reach** | Outlet's audience size + their byline traction | 2x |
| **Engagement** | Do they respond to pitches publicly / on X? | 2x |
| **Recency** | Have they written about a related topic in last 30d? | 1x |
**Tiering:**
- **Tier 1 (8–10):** Personal pitch with original angle. Custom each time.
- **Tier 2 (5–7):** Standard pitch, lightly customized.
- **Tier 3 (below 5):** Skip or pitch only when story is exceptional.
---
## Pitch Templates by Angle
Six starting structures to test. Adapt one to the verified story, recipient
preferences, and outlet format.
### 1. Data story
```
Subject: [Specific stat]: [implication]
Hi [name],
I noticed you covered [recent article]. Wanted to share data that might
be relevant.
We [analyzed N / surveyed N / tracked N] and found:
• [Stat 1 with surprise factor]
• [Stat 2]
• [Stat 3]
The most interesting pattern: [one-sentence insight].
Full data + methodology here: [link to one-pager, not your homepage]
Happy to share the raw dataset, jump on a call, or connect you with
[customer who's relevant].
[your name + 1-line credential]
```
### 2. Exclusive launch / milestone
```
Subject: Exclusive: [specific milestone] at [company]
Hi [name],
I have an exclusive on [milestone] that I think fits your [beat] coverage.
The story: [one sentence]
Why it matters: [one sentence: for their readers, not for you]
What's new: [the actual news, not the marketing line]
Embargo until [day, time, timezone]. Would love to give you first
window. Press kit + assets: [link]
Free to talk [two specific time options].
[your name]
```
### 3. Op-ed / contributed piece
```
Subject: Op-ed pitch: [provocative thesis]
Hi [name],
I read your piece on [recent article]: sharp take on [specific point].
I'd like to pitch a 700-word op-ed: "[Thesis as a headline]"
Core argument:
• [Point 1]
• [Point 2]
• [Point 3: the surprising one]
Why me: [1 sentence: credential or unique vantage]
Why now: [1 sentence: the news hook]
Can have a draft to you by [date]. Happy to adapt to your house style.
[your name]
```
### 4. Customer story
```
Subject: Customer story for [their beat]: [specific outcome]
Hi [name],
For your [beat] coverage, I have a [customer type] willing to talk on
the record about [specific outcome].
The hook: [customer] [did something specific] and [measurable result].
The interesting part: [the surprising or counterintuitive detail].
Customer details:
• Name: [name, title, company]
• Available: [windows]
• Willing to share: [data points / screenshots / metrics]
Happy to coordinate the intro.
[your name]
```
### 5. Trend piece / connector
```
Subject: Trend forming in [space]: three signals
Hi [name],
Three things in [space] this month that I think connect:
1. [Signal 1 with link]
2. [Signal 2 with link]
3. [Signal 3: yours, briefly]
The pattern: [one sentence].
This might be early for a piece, but if you're tracking the space I
wanted to flag it. Happy to share data we've collected or connect you
with others seeing the same.
[your name]
```
### 6. Newsjack response
```
Subject: Re: [their article headline]: quick data point
Hi [name],
Saw your piece on [story] this morning. Wanted to add a relevant
data point in case you do a follow-up.
[One-sentence stat or insight].
Source: [our data / our customers / our analysis]
Methodology: [one sentence]
Quotable: "[a sentence you'd be comfortable seeing in print]"
If useful for a follow-up, I'm around all day at this number: [phone].
[your name]
```
---
## Subject Lines That Get Opened
The subject line affects comprehension and may affect opens. Treat these as
drafting hypotheses and verify outlet preferences where published:
- **Concise enough for the current inbox preview**: verify the relevant client
or test multiple readable lengths
- **Lead with the specific**: "73% of devs deploy to prod on Fridays" beats "New data on developer workflows"
- **Promise a story, not a product**: "Why [trend]" beats "[Company] launches [thing]"
- **Use prefixes that signal value**: "Exclusive:", "Data:", "Op-ed pitch:", "Re: [their article]"
**Test against this question:** would *you* open this in a 200-email inbox?
**Patterns to test:**
- "[Specific stat]: [implication]": "73% of agents fail this test"
- "Exclusive: [milestone]": "Exclusive: Anthropic launches AgentOS"
- "Re: [their headline]": direct response to recent coverage
- "[Provocative thesis]": "Why VC funding is bad for AI safety"
**Patterns that get deleted:**
- "Press release: [boring]"
- "[Company] announces [thing]"
- "Story idea for you!"
- "Following up on my previous email"
- Any subject line with "innovative," "disruptive," "revolutionary"
---
## Voice and Structure
### Length
Start under 200 words, then fit the verified outlet preference and story
complexity. In Muck Rack's State of Journalism 2026 survey, 69% of journalists
said they prefer pitches under 200 words, 62% prefer a one-to-one email over
other channels, and 78% prefer pitches sent before noon in their time zone.
Treat these as defaults, not laws; an individual journalist's stated preference
wins. Source: [Muck Rack, State of Journalism 2026](https://muckrack.com/resources/research/state-of-journalism).
### Structure
1. **One-line context**: why you're emailing them specifically (their recent article, their beat)
2. **The story**: what it is, in one sentence
3. **Why it matters to their readers**: not why it matters to you
4. **Proof**: data, customer, quote, link
5. **The ask**: interview, embargo, quote, link
### Voice
- Sound like a person, not a press release
- Reference their actual recent work: proves you read them
- Don't use emoji unless they do
- Don't open with "I hope this finds you well": burn it
- Don't ask "did you get my email?" follow-ups (see [Follow-up](#embargoes-exclusives-and-follow-up-etiquette))
### Banned vocabulary
Revolutionary, disruptive, game-changing, paradigm shift, leverage, synergy, robust, seamless, holistic, world-class, best-in-class, next-generation, cutting-edge, AI-powered (unless that's the actual differentiation), at-the-end-of-the-day.
---
## Embargoes, Exclusives, and Follow-Up Etiquette
### Embargoes
An embargo is "you can write this story, but don't publish until [time]."
- Offer an embargo only after current outlet practice, relationship, disclosure,
and approval checks support it
- State the embargo time clearly: day, time, timezone
- If they break embargo, your relationship with them is over
### Exclusives
"Only you get this story": powerful tool, use sparingly.
- Evaluate exclusivity from audience fit, story value, terms, and opportunity
cost rather than outlet tier alone
- **Be honest about scope**: "exclusive to [outlet] in the US" is fine
- **Have a parallel plan**: what you publish/pitch the next day after the exclusive runs
### Follow-up cadence hypothesis
- Start from any preference or deadline published by the journalist or outlet.
- If no preference exists, propose at most one follow-up only when it adds
material new information; choose timing from story velocity and prior
response data.
- Stop when the deadline, relevance, or consent context closes. Re-pitch only
with a genuinely new, verified story.
Every initial pitch and follow-up remains a separate draft until the user
explicitly approves the exact recipient, identity, channel, and content.
**Never:**
- "Bumping this up" / "Did you see my email?"
- Multi-day silent follow-ups with no new value
- Same pitch reformatted
---
## Pitch Killers
Things that instantly disqualify your pitch:
- Wrong name / wrong outlet (autoreplace fail)
- Pitching topics they explicitly don't cover
- Press release attached as PDF (just paste the key bits)
- Long signature with logos and disclaimers
- CC'ing 5 other journalists on the same email
- "Per my last email" energy
- Asking them to sign an NDA before talking
- Pitching a story you can't actually tell (no customer willing to talk, no data ready to share)
---
## Tooling
### Finding journalist contact info
Use only a publicly listed professional email, the outlet's official contact
path, or a contact database the user is licensed and authorized to use. Discover
whether a current callable verifier exists before relying on one; otherwise ask
the user to verify manually. Do not guess addresses, send bounce probes, or
present an inferred address as confirmed.
### Researching their recent work
First discover whether an authorized browser or source reader is currently
callable and confirm the visible account identity and source terms. If
available:
- Review a bounded recent sample from the journalist's outlet author page and
record article headlines, dates, and links.
- Review public professional posts only when the platform permits access.
- Confirm the current role from the outlet profile or another first-party
professional source.
If no browser is available, ask the user for the author-page URL, article links,
or an export and provide a manual research checklist. Never claim the research
was performed when only the checklist was produced.
Output what you find as:
```
JOURNALIST PROFILE: [name]
Outlet: [name]
Beat: [topics from bounded recent sample]
Recent angle: [pattern you noticed]
Recent X activity: [what they're posting]
Score: [X/40 from rubric]
Best pitch angle: [from template library]
Email: [confirmed]
```
### Maintaining the media list
Store in `.agents/media-list.md` or a user-approved CSV. Recheck the role,
outlet, contact path, and source date immediately before each pitch; use a
calendar-based refresh only if the user's send cadence supports it.
```markdown
## Highest-evidence current tier
| Name | Outlet | Beat | Last contact | Last coverage | Email | Score |
|------|--------|------|--------------|---------------|-------|-------|
| ... | ... | ... | 2026-05-15 | none yet | ... | 9/10 |
```
### Pitch tracking
Track in a simple spreadsheet:
- Date sent
- Subject line
- Journalist
- Outlet
- Response (open / reply / pass / coverage)
- What they said
Review results after a predeclared, capacity-bounded cohort. Do not claim a
minimum sample is sufficient; report uncertainty and expand only when another
cohort is justified.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |