audit-before-i-tweet
Review Your Tweet Before Publishing
Short tweets can lose context, overstate claims, or expose identifiable people. This skill checks a finished draft and ranks up to three fixes before publishing.
Installer avec mon Agent
Copiez cette demande dans votre Agent. Elle contient la page Skill de référence et le manifeste.
Review the Skillstore skill "audit-before-i-tweet" from https://skillstore.io/skills/prateeks367-audit-before-i-tweet.md and its manifest at https://skillstore.io/api/skills/prateeks367-audit-before-i-tweet/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Votre Agent doit toujours présenter son plan et demander toute confirmation exigée par la politique de sécurité.
Ressources lisibles par les agents
Utilisez ces liens lorsqu’un Agent IA, un crawler ou un script a besoin d’un contexte propre au lieu de lire toute la page.
Tester
Utilisation de « audit-before-i-tweet ». Ready to post: Our tool cuts support costs by 50% for everyone. Evidence: one four-week customer pilot.
RĂ©sultat attendu:
- Fix 1: Replace 'for everyone' with the scope of the single customer pilot.
- Fix 2: Name the source for the 50% figure and verify how support costs were measured.
- Clean check: The subject is clear without surrounding context.
- Decision: Revise the claim before posting.
Utilisation de « audit-before-i-tweet ». Final review: We moved our weekly planning meeting from Monday to Tuesday. For our team, that leaves Monday free for focused work.
RĂ©sultat attendu:
- Clean checks: The subject is clear, and 'for our team' preserves the claim's scope.
- Clean checks: No numerical result or identifiable outside person appears in the draft.
- Voice: No reference posts were supplied, so personal voice consistency remains unverified.
- Decision: I would post this as written.
Audit de sécurité
SûrAll five static findings are false positives: four identify Markdown tool names, and one misclassifies editorial guidance as system reconnaissance. No evidence found of command execution, credential access, exfiltration, or prompt injection in SKILL.md. Optional VoiceMOAT publishing is described with a preview and one-time confirmation; its implementation is not included.
Facteurs de risque
⚙️ Commandes externes (4)
Partager et citer ce rapport
Partagez le rapport d’évaluation versionné, le badge neutre, la carte intégrable et les citations. Skillstore présente les preuves sans décider si ce Skill est sûr.
Copier le lien du rapport
https://skillstore.io/skills/prateeks367-audit-before-i-tweet/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportBadge Markdown
[](https://skillstore.io/skills/prateeks367-audit-before-i-tweet?utm_source=security_passport_badge)Badge HTML
<a href="https://skillstore.io/skills/prateeks367-audit-before-i-tweet?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/prateeks367-audit-before-i-tweet/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Carte intégrable
<iframe src="https://skillstore.io/embed/skills/prateeks367-audit-before-i-tweet.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Citations académiques (APA · BibTeX · CFF)
Citation APA
prateeks367. (2026). audit-before-i-tweet security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/prateeks367-audit-before-i-tweet/audits/1Citation BibTeX
@techreport{prateeks367-prateeks367-audit-before-i-tweet-2026,
author = {prateeks367},
title = {audit-before-i-tweet security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/prateeks367-audit-before-i-tweet/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "audit-before-i-tweet security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "prateeks367"
date-released: "2026-10-02"
url: "https://skillstore.io/skills/prateeks367-audit-before-i-tweet/audits/1"
identifiers:
- type: other
value: "skillstore:prateeks367-audit-before-i-tweet:audit:1"
description: "Skillstore immutable audit report identifier"
Score Skillstore
Pourquoi ce score Niveau de confiance des preuves: MoyenCe que vous pouvez créer
Check a product announcement
Review an announcement for unclear subjects, unsupported numbers, and missing qualifiers before posting.
Review a client-related anecdote
Check whether readers could identify a client or misinterpret a professional story.
Prioritize final editorial fixes
Review a finished brand tweet, preserve its voice, and return no more than three prioritized changes.
Essayez ces prompts
Audit this finished tweet before I post it: [draft]. Rank up to three fixes and say whether you would post it unchanged.
Review [draft] for unsupported claims and identifiable people. Use [sources] as context, and tell me what still needs verification.
Audit [draft] as a standalone screenshot. Compare it with [full thought] to find unintended readings or lost qualifiers. Rank only necessary fixes.
Audit [draft] against [my published posts]. If VoiceMOAT is connected, use its voice profile and scoring. Rank fixes; do not publish without confirmation.
Bonnes pratiques
- Provide a finished draft and supporting sources for numbers or factual claims.
- Supply the original thought when shortening may have removed important qualifiers.
- Use your own published posts as voice references and review every suggested change before publishing.
Éviter
- Requesting a new tweet from an empty brief instead of submitting a completed draft.
- Treating the review as a performance forecast or guaranteed factual verification.
- Rewriting the entire tweet when a small correction would resolve the issue.
Foire aux questions
Does this skill write tweets from scratch?
What does the audit check?
How many fixes will I receive?
Will it predict engagement or give a score?
Do I need VoiceMOAT?
Does it publish automatically?
Détails du développeur
Auteur
prateeks367Licence
MIT
RĂ©vision Skillstore
r1
Avis de version
L’auteur n’a déclaré aucune version.
RĂ©f
5726d703fd79596f46a4232d911221f962d9af1d
RĂ©cence de maintenance
02/10/2026
Utilisation
0 téléchargements · 0 vues
Structure de fichiers
đź“„ SKILL.md