Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsCompany Voice System
Maintain a persistent, plain-file editorial memory to generate brand-accurate content and strategy.
Free
Company Voice System
Example session with this skill installed
Create a primary voice file for a technical consultancy that values depth over hype, then write a short LinkedIn post about our new cloud migration service using the 'confirmed' status.
- Read your context and instructions
- Compiled the company voice system
I have initialized the editorial memory and drafted the content based on the new voice rules.
Editorial Memory & Content Draft
---
id: voice.primary
status: confirmed
source: user-instruction
updated: 2024-05-20
depends_on:
- positioning.technical-consultancy
affects:
- style. LinkedIn-post
---
# Primary Voice: Technical Depth
Our voice prioritizes evidence and technical accuracy over marketing superlatives. We use precise terminology and avoid "game-changing" or "seamless" descriptors. Tone is authoritative, calm, and peer-to-peer.
Generated Content: LinkedIn Post
Target Audience: CTOs, Cloud Architects
Content Pillar: Service Launch (Cloud Migration)
Tone: Authoritative / Technical
"Cloud migration often fails at the data persistence layer, not the application layer. Our new migration service focuses on stateful workload transitions and latency benchmarking rather than generic lift-and-shift tactics. We have documented our approach to handling complex schema refactoring during zero-downtime cutovers. Read the technical brief below."
Next steps
- Review the
voice.primaryfile in.editorial-voice/to ensure the "depth over hype" constraints are accurate. - Run
scripts/validate_editorial_memory.pyto confirm the new file is correctly indexed. - Provide feedback on the post to refine the
style.LinkedIn-postprovisional rules.
Connects securely to your tools. The creator never sees your data.
About this skill
AI content generation often defaults to a generic, robotic tone that fails to capture a brand's specific positioning or values. Without a persistent record of voice rules, every new prompt requires re-explaining the audience, tone, and constraints. This leads to inconsistent output and significant manual editing.
This skill implements a persistent plain-file knowledge system to maintain an organization's editorial memory. It moves beyond simple prompting by using an adaptive discovery process to build a structured foundation of confirmed, inferred, and provisional brand rules.
What it does
- Adaptive discovery conducts targeted interviews to capture positioning, audience needs, and tone without long questionnaires.
- Editorial memory stores voice, claims, and strategy in a structured, linked file system for long-term consistency.
- Contextual routing loads only the specific files needed for a task, preventing context window bloat.
- Claim safety reviews every piece of generated content against stored rules to flag blocking issues or inaccuracies.
- System maintenance handles updates to audience or product pillars while preserving unrelated decisions and tracking dependencies.
How it works
- Discovery: Initialize the system by answering high-value questions about your organization and goals.
- Foundational build: The skill generates an
.editorial-voice/directory with specific metadata-tagged Markdown files. - Execution: Request content generation; the agent creates a plan based on stored pillars before writing.
- Validation: Run the provided Python validation script to ensure links and decision statuses remain consistent.
Frameworks & tools
The skill utilizes a structured Markdown file system with YAML frontmatter. It includes a Python validation script validate_editorial_memory.py for maintaining repository integrity.
Why this beats prompting it yourself
One-off prompts lack memory, forcing you to paste style guides repeatedly. This skill builds a durable repository that evolves with your company, ensuring the agent "remembers" your positioning, prohibited claims, and preferred rhythm across every session.
Use cases
- Building a multi-channel content backlog for a new product launch.
- Enforcing strict claim safety and tone for a regulated consultancy.
- Transitioning brand voice from "corporate" to "community-driven" across all assets.
- Managing search strategy and topic pillars for a technical blog.
Known limitations
Requires a workspace capable of file system operations to store the editorial memory. Python is required only if you wish to run the automated consistency validation script.
What's included (2 skills)
company-voice-skill
Route requests to the Company Voice skill for discovering, maintaining, and applying an organization's editorial voice and content strategy. Use when company-specific content depends on a persistent editorial memory; the detailed workflow lives in the nested skill folder.
company-voice
Build, store, maintain, and apply a company's editorial voice and content strategy as a persistent plain-file knowledge system. Use when Codex is asked to discover an organization, define positioning or content strategy, create or update a brand/editorial voice memory, generate company content from stored voice rules, review content for voice and claim safety, manage a content backlog, or evolve publishing/performance rules over time.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scanned
Verified clean 3 days ago
- Free to download with an account
Requires
Brownsmithdynamics