Sync tested skill from gingiris-skills 9deb2a5
Browse files
SKILL.md
ADDED
|
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
name: devrel-playbook
|
| 3 |
+
description: Build and measure Developer Relations programs across documentation, community, technical content, integrations, events and developer advocacy. Use when designing DevRel strategy, improving developer activation, launching an API/SDK, evaluating sponsorships, or connecting community activity to product adoption and revenue.
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# DevRel Playbook — Developer Activation, Community and Revenue
|
| 7 |
+
|
| 8 |
+
DevRel exists to reduce the distance between a developer's first exposure and sustained product value. Do not optimize only for members, event attendance or impressions.
|
| 9 |
+
|
| 10 |
+
## Diagnose the bottleneck
|
| 11 |
+
|
| 12 |
+
Choose one primary problem:
|
| 13 |
+
|
| 14 |
+
| Problem | Evidence | First intervention |
|
| 15 |
+
|---|---|---|
|
| 16 |
+
| Discovery | ICP developers do not know the project | technical content and ecosystem distribution |
|
| 17 |
+
| Understanding | Docs traffic is high, quickstart completion is low | README/docs usability tests |
|
| 18 |
+
| Activation | Keys/SDK installed, first successful call is low | sample app and error-path repair |
|
| 19 |
+
| Retention | Developers try once and disappear | use-case education, office hours, lifecycle prompts |
|
| 20 |
+
| Contribution | Users want changes but do not contribute | issue design, maintainer SLA, contributor onboarding |
|
| 21 |
+
| Enterprise pull | Teams use product but procurement is invisible | usage-qualified account handoff to sales |
|
| 22 |
+
|
| 23 |
+
## Developer journey instrumentation
|
| 24 |
+
|
| 25 |
+
Track:
|
| 26 |
+
|
| 27 |
+
```text
|
| 28 |
+
source → docs/README visit → quickstart start → first success → second session → production use → contribution/qualified account
|
| 29 |
+
```
|
| 30 |
+
|
| 31 |
+
Define each event and cohort window. Separate employees, bots, test accounts and event attendees who never touched the product.
|
| 32 |
+
|
| 33 |
+
## Documentation system
|
| 34 |
+
|
| 35 |
+
Maintain four layers:
|
| 36 |
+
|
| 37 |
+
1. **README**: outcome, demo, quickstart, license and support path;
|
| 38 |
+
2. **Tutorial**: one job completed end to end;
|
| 39 |
+
3. **How-to**: focused operational tasks and integrations;
|
| 40 |
+
4. **Reference**: complete APIs, errors, limits and versions.
|
| 41 |
+
|
| 42 |
+
Run five-developer task tests quarterly. Record completion time, failure step, search terms and recovery success. Documentation is a product surface, not a publishing queue.
|
| 43 |
+
|
| 44 |
+
## Community operating model
|
| 45 |
+
|
| 46 |
+
- Assign owner and response SLA for support, bugs and feature discussions.
|
| 47 |
+
- Separate announcements, help, showcase and contributor channels.
|
| 48 |
+
- Turn repeated questions into docs; turn reproducible bugs into issues.
|
| 49 |
+
- Recognize substantive help, not message volume.
|
| 50 |
+
- Publish moderation and escalation rules before growth.
|
| 51 |
+
|
| 52 |
+
Weekly review: unanswered questions, median time to useful answer, activated community members, recurring friction and contributions merged.
|
| 53 |
+
|
| 54 |
+
## Technical content and comparison assets
|
| 55 |
+
|
| 56 |
+
Prioritize content a developer can verify:
|
| 57 |
+
|
| 58 |
+
- reproducible benchmark with methodology;
|
| 59 |
+
- architecture/deep-dive explaining tradeoffs;
|
| 60 |
+
- migration or integration tutorial;
|
| 61 |
+
- honest comparison page stating where each option wins;
|
| 62 |
+
- user build story with repo or demo.
|
| 63 |
+
|
| 64 |
+
The anonymized Apache-ecosystem campaign in the Gingiris case library combined README repair, comparison content and backlinks/sponsor distribution; it reported 200K impressions and +401 stars in 10 days. Treat this as historical evidence, not guaranteed lift.
|
| 65 |
+
|
| 66 |
+
## Events, hackathons and sponsorships
|
| 67 |
+
|
| 68 |
+
Approve only when the event reaches the target developer and has a post-event activation path.
|
| 69 |
+
|
| 70 |
+
Before: define build prompt, sample app, mentor coverage, attribution and success event.
|
| 71 |
+
During: measure builders who reach first success, not registrations.
|
| 72 |
+
After: route viable projects to showcase, contributor or customer tracks and measure D7/D30 continuation.
|
| 73 |
+
|
| 74 |
+
Event scorecard:
|
| 75 |
+
|
| 76 |
+
```text
|
| 77 |
+
qualified registrants | builders started | first success | demos completed | D7 active | contributions | opportunities | total cost
|
| 78 |
+
```
|
| 79 |
+
|
| 80 |
+
## Ecosystem and backlink program
|
| 81 |
+
|
| 82 |
+
- Maintain official integration pages and reciprocal technical documentation.
|
| 83 |
+
- Contribute useful examples to ecosystem repositories before requesting promotion.
|
| 84 |
+
- Use sponsor/newsletter placements only with source-tagged links and a relevant developer offer.
|
| 85 |
+
- Reject paid link schemes and irrelevant directory volume.
|
| 86 |
+
|
| 87 |
+
## DevRel-to-sales boundary
|
| 88 |
+
|
| 89 |
+
DevRel educates and earns trust; it does not disguise sales outreach as community help. Hand off an account only when product usage, team expansion, security/procurement questions or explicit intent creates a qualified signal. Tell the developer when a commercial teammate is joining.
|
| 90 |
+
|
| 91 |
+
## 30-day operating plan
|
| 92 |
+
|
| 93 |
+
- Week 1: baseline journey, interview five developers, identify one bottleneck.
|
| 94 |
+
- Week 2: repair the highest-frequency docs/product failure and ship one proof asset.
|
| 95 |
+
- Week 3: distribute through two relevant communities or ecosystem partners.
|
| 96 |
+
- Week 4: compare activation/retention against baseline and decide keep, change or stop.
|
| 97 |
+
|
| 98 |
+
## Required output
|
| 99 |
+
|
| 100 |
+
Return a developer journey, bottleneck diagnosis, 30-day plan, channel owners, measurement schema, community escalation rules and a monthly executive report connecting activity to activated/retained developers.
|
| 101 |
+
|
| 102 |
+
## Compliance
|
| 103 |
+
|
| 104 |
+
Never buy stars, fake community activity, conceal sponsorship, scrape private member data or manufacture benchmarks. Obtain consent before using developer stories or code.
|