Advocacy Program Designer
Design voluntary, disclosed employee and founder sharing programs without manufactured engagement.
The skill specifies minimal roster fields, default voluntary participation, operation: propose writes, confirmation before saving, and no posting, engagement, or DM automation. It discloses data flows, manual delivery for closed platforms, and sole-writer registry boundaries. Four points are deducted for the lack of a complete rollback plan, unverifiable runtime/registry dependencies in the supplied evidence, and legal disclosure references that are named but not auditable here.
Inputs, outputs, done conditions, abnormal branches, and termination rules are described, including assigned-mode risks, missing opt-in evidence, paid conversion, and platform access classes. Two points are deducted because the cited files, registry-events.py, and connectors were not supplied for reproduction, and no skill-specific tests or standardized failure-message format are shown. Static calibration caps this at 10.
Triggers, audience, argument hint, opt-in/assigned modes, and non-fit boundaries are clear, with manual handling and Chinese safety guidance for Chinese platforms. Four points are deducted because platform-native disclosure, availability, and regional differences are not detailed per platform; Slack/Teams may not fit every mainland-China team, and environment-fit evidence is limited.
Frontmatter is substantial, and the document is layered through Quick Start, Skill Contract, Data Sources, Instructions, Save Results, and Next Best Skill. Apache-2.0, versioning, and dependency boundaries are stated, with repository governance context in README and SECURITY.md. Five points are deducted for missing full output examples, FAQs, changelog, explicit maintainer responsibility, and update path, plus assumptions hidden in unavailable referenced documents.
The expected deliverable, inputs, done conditions, and handoff are concrete. Per-person variation, disclosure lines, staggered windows, anti-pod guardrails, and human-delivery packaging are directly useful. Nine points are deducted because there is no representative output or third-party execution evidence, compliance results still require human review, and registry/save behavior depends on unsupplied runtime components. Static calibration caps this at 7.
The skill names ECHO, FTC guidance, the Chinese advertising regulation, and repository protocols, and distinguishes user-provided, measured, and source-needed claims. Seven points are deducted because the supplied material contains no specific cited excerpts, independent reproduction record, or skill-path test coverage; repository tests cover general architecture and audit validation rather than this skill's behavior. Static calibration caps this at 5.
- Disclosure compliance and platform rules are only summarized; review each item with qualified counsel or policy owners before publishing.
- Do not treat a manager assertion as opt-in evidence, and do not bypass operation: propose or authorized registry boundaries for person-record writes.
- Several referenced documents, runtimes, and connectors were not supplied; do not assume writes, projections, or platform checks will succeed until those dependencies are verified.
- Slack/Teams and some overseas platforms' reachability, native disclosure support, and mainland-China compliance are not sufficiently demonstrated by this material.
What does this skill do, and when should you use it?
This skill designs employee-advocacy and founder-led sharing programs, defaulting to voluntary opt-in while supporting assigned mode with explicit risk flags. It produces a program blueprint, a minimal advocate-roster specification, per-person share kits, staggered posting windows, disclosure lines, and Slack/Teams distribution guidance. The workflow requires people to post in their own words on their own schedules and prohibits identical coordinated reshares, engagement rings, and automation. It does not publish content, run recruitment conversations, manage paid creators, or perform the ECHO audit.
Reads the user-provided program goal, participant list, mode preference, platforms, source post or asset, and approved claim wording, plus authorized roster and pending event data; decides between opt-in and assigned mode and flags coercion, authenticity, and disclosure risks; creates varied share-kit angles and fill-in-your-own-words structures for each advocate; schedules advocate-chosen posting windows across 3–7 days; states guardrails against identical reshares, engagement rings, mandated like/comment rounds, automated replies, and quotas; specifies Slack/Teams distribution, opt-out handling, and lightweight attribution; submits roster and claim candidates through the authorized registry-event proposal path rather than writing canonical registry files directly.
- A marketing lead at a 40-person developer-tools company needs an opt-in advocacy program for LinkedIn and Bluesky.
- A founder wants a weekly sharing kit for product updates with individualized angles and disclosure language.
- Leadership wants every employee to reshare a launch post at the same time and needs the plan documented with its compliance and platform risks.
- A team has 12 opted-in advocates and needs a varied kit, disclosure lines, and staggered windows for a changelog post.
What are this skill's strengths and limitations?
- Defaults to voluntary participation and requires evidence of the voluntary basis.
- Requires per-person variation and an individual disclosure line in every share kit.
- Clearly prohibits coordinated identical reshares, engagement rings, automated replies, and quotas.
- Covers Slack/Teams distribution, platform access classes, and 3–7 day staggered windows.
- Only designs blueprints and share packages; it does not automate posting, engagement, or DMs.
- LinkedIn, Instagram, TikTok, X, and several Chinese platforms rely mainly on user exports or manual-package deliverables rather than automated publishing.
- Users must provide real participants, source assets, and approved product claims; unsupported claims are marked `[needs source]`.
- Roster and claim writes depend on the authorized `registry-events.py` proposal flow and cannot directly update canonical registry files.
How do you install this skill?
Install the repository in Claude Code or another Agent Skills-compatible host, then enable or load social/craft/advocacy-program-designer/SKILL.md. The README documents Claude Code installation with /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, followed by /plugin install aaron-marketing@aaron; generic hosts can use npx skills add aaron-he-zhu/aaron-marketing-skills.
How do you use this skill?
Use a prompt such as: Design an opt-in employee advocacy program for our 40-person dev-tool company — LinkedIn + Bluesky, founder posts weekly. You can also specify opt-in or assigned, the team size or advocate list, and target platforms. Route paid creator work, one-to-one recruitment, or ECHO review to the repository's campaign-planner, outreach-manager, or social-quality-auditor skills.
How does this skill compare with similar options?
Unlike campaign-planner, which covers paid creator campaigns, this skill is for employee and founder-led sharing. Unlike outreach-manager, it does not handle one-to-one invitations, follow-ups, or objection handling.