What does this skill do, and when should you use it?
Social Quality Auditor is the organic-social audit skill in Aaron Marketing Skills, using the ECHO framework. It supports two separate modes: auditing a specific social asset or batch, and profiling the maturity of a community, B2C, or founder-led channel program. It reads an asset package or channel portfolio, governance projections, and dated evidence, then produces a structured audit artifact without publishing content, resuming queues, or changing canonical records. Missing evidence is reported as Unknown and can result in NEEDS_INPUT/UNDECIDED rather than an automatic failure.
Declares the target, profile, assessment mode, program archetype, channels, market, window, and observation date; runs the ECHO asset gate or a program-maturity profile; checks channel truth, claims, applicable disclosures, engagement manipulation, UGC republication permission, and denominator integrity; verifies controls including ECHO-E1, C1, C2, H1, H2, and O1; records source, date, evidence type, and confidence for observed states; and reports status, verdict, score or coverage, Unknown inputs, critical controls, and fixes. With explicit authorization, it can persist and validate a complete audit artifact in the designated memory path.
- A social team needs to clear a multi-post package before it enters the publishing queue.
- A paused content queue is about to resume, but channel ownership, disclosures, or UGC permissions are uncertain.
- A community manager wants a maturity baseline for the last 90 days with outcomes reported separately.
- A B2C social lead needs a feed-led program assessment rather than a review of one post.
- A founder-led social program needs an evidence-based maturity rerun with missing private inputs made explicit.
How do you install this skill?
- Key referenced files and runtime components were not included, so the skill's complete runnable behavior cannot be confirmed statically.
- Actual scoring depends on authorized current channel, claim, permission, and observation evidence; missing private evidence must remain Unknown and cannot support automatic publish approval.
- Verify that the skill's 18.0.0 version, the 17.x support statement in SECURITY.md, maintenance ownership, and update path are aligned.
- Shell / CLI
- Network access
- Local filesystem
GitPython standard-library validator and scorer scripts
This skill is distributed in the Aaron Marketing Skills repository and is compatible with Claude Code and compatible Agent Skills hosts. In Claude Code, run: /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then: /plugin install aaron-marketing@aaron. For generic hosts, run: npx skills add aaron-he-zhu/aaron-marketing-skills; alternatively, clone the repository with: git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The source documents collection-level installation; it does not document a separate standalone installation command for this skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/social/host/social-quality-auditor" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Run ECHO asset-gate on this five-post batch against channel/claims/UGC records.
- Audit the last 90 days with program-maturity-community; report outcomes separately.
Place the skill where the host discovers Agent Skills, then use a prompt such as: "Run ECHO asset-gate on this five-post batch against channel/claims/UGC records." Or: "Audit the last 90 days with program-maturity-community; report outcomes separately." Use asset-gate for an asset review, or exactly one of program-maturity-community, program-maturity-b2c, or program-maturity-founder for a maturity review. Never combine asset and program profiles into one score, and do not publish or unpause a queue solely because of the audit.
What are this skill's strengths and limitations?
- Separates publish-package review from operating-program maturity scoring.
- Covers concrete controls for channel truth, claims, disclosures, engagement manipulation, UGC rights, and denominators.
- Treats missing private evidence as Unknown instead of converting it into an automatic failure.
- Requires provenance, dates, confidence, and structured ECHO results for auditability.
- Apache-2.0 licensed and designed for Claude Code and compatible Agent Skills hosts.
- It does not approve or execute publishing, queue resumption, or registry updates.
- Closed-platform evidence requires a user-provided native export, which can limit coverage.
- If the referenced runbook, scoring materials, catalogs, or validator are unavailable, it must return NOT_SCORED.
- The source provides no standalone test-coverage report or per-platform validation results for this skill.
- Business outcomes remain outside the rubric, so the skill does not produce a single combined performance score.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Social Quality Gate this page | 62 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Brand Voice Dossier Builder | 59 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Creator Content Auditor | 56 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Advocacy Program Designer | 59 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Newsletter Monetization Planner | 61 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
Unlike creator-content-auditor, this skill audits social asset batches or channel-program maturity rather than creator deliverables. Unlike launch-readiness-auditor, it does not assess product-launch readiness.
How did FollowSkills review this skill?
The skill explicitly limits reads to asset packages, channel portfolios, governance projections, and dated evidence; it writes only permissioned audit artifacts and forbids publishing, queue, registry, or engagement side effects. It also specifies permission, UGC-rights, sensitive-state, and fail-closed principles. Five points are deducted because the referenced runtime, connector boundaries, and concrete confirmation/rollback implementation are not fully present, and publisher identity is unverified.
It defines a single declared profile, explicit unknown handling, typed scoring, fail-closed behavior when runtime components are unavailable, and validation before persistence. The key reference files, scorer, and validator were not all supplied, and no reproduction was executed, so the static score is capped at 9 with one point deducted for unresolved implementation uncertainty.
Triggers, the asset gate, three maturity profiles, input scope, and neighboring non-fit skills are clearly described, with Chinese-language framing. Repository context states that a keyless Tier 1 path exists and the core function does not depend entirely on overseas services. Three points are deducted because platform, market, abnormal-input, and precise semantic-trigger boundaries have limited evidence.
The skill has structured metadata, versioning, an Apache-2.0 license, quick-start examples, data-source guidance, limitation disclosures, report rules, and next-skill handoffs. Four points are deducted because no skill-specific changelog, explicit maintenance owner, or complete installation/troubleshooting path is supplied; SECURITY.md also still lists 17.x as the supported line while the skill is version 18.0.0.
The intended output is a reviewable ECHO asset or program-maturity result, with defined outputs, unknowns, vetoes, and side-effect prohibitions; the worked examples clarify expected decisions. Nine points are deducted because no representative executed output or independent execution evidence is supplied, and key reference materials are absent, leaving only limited static support for direct usability.
The supplied files include an immutable runtime snapshot, source digest, explicit evidence types, and some repository regression tests, providing partial traceability. Six points are deducted because the tests mainly cover general architecture and artifact validation rather than end-to-end paths for this skill, and this review did not execute anything.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →