Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
Muse
OpenClaw+21 moreClaude CodeCursorCodex CLIMuseOpenClaw
Manus
Grok
Grok Bot+18 agentsRedactFirst: Pre Publish IP Decoupling Guard
Anonymizes code, configs, and prompts to prevent IP leaks and NDA violations before public release.
Free
RedactFirst: Pre Publish IP Decoupling Guard
Example session with this skill installed
Sanitize this internal database config for a blog post:
Acme Corp staging
host: db.internal.acme.co
user: dev_admin
pass: sk_live_55a92b
region: US-East-1 (Virginia)
- Read your context and instructions
- Compiled the redactfirst
- Generated the document
Scan complete. Status:
BREACH. Several direct identifiers and live secrets were detected. You must rotate the API key at the provider before publishing this content.
RedactFirst Sanitization Report
| Category | Finding | Risk | Action |
|---|---|---|---|
| Direct Identifier | Acme Corp | High | Replaced with Client-A |
| Internal Domain | db.internal.acme.co | High | Replaced with db.vendor.example |
| Secret Survivor | sk_live_55a92b | Critical | ROTATE KEY IMMEDIATELY |
| Indirect Identifier | US-East-1 (Virginia) | Medium | Replaced with region-1 |
Sanitized Artifact
# Client-A staging
host: db.vendor.example
user: user_admin
pass: "[REDACTED_SECRET_ROTATE_AT_SOURCE]"
region: region-1
Next steps
1.
Rotate Secrets: The detected key sk_live_55a92b is a live credential. Redacting it in the text does not invalidate the leak.
2.
Verify Placeholders: Ensure Client-A and region-1 are used consistently across all other files in this blog post.
3.
Check Logic: Confirm the sanitized YAML still accurately demonstrates the configuration logic intended for your readers.
redactfirst-pre-publish-ip-decoupling-gu.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Publishing proprietary code or client projects to public repositories often leads to accidental IP leaks, NDA violations, and credential exposure. This skill automates the decoupling of intellectual property from your codebase, ensuring that what you share is clean, anonymous, and safe for public consumption.
What it does
- Identifier Redaction scans for and replaces client names, internal domains, and personal data with consistent placeholders.
- Secret Detection identifies API keys, tokens, and hashes that require rotation rather than simple redaction.
- Structural Integrity preserves all code logic and flow while stripping only the identifying metadata and strings.
- Indirect Mapping flags unique tech stacks or dated comments that could be used to re-identify the source through context.
- Risk Scoring provides a categorical verdict (CLEAR, WARNING, BREACH) based on the remaining footprint.
How it works
- Direct Scan lists all occurrences of company names, private URLs, and employee credentials.
- Neutral Mapping replaces those values with consistent, non-identifying strings like
client-aorvendor.example. - Indirect Flagging highlights project-specific naming conventions or regional references for manual review.
- Final Reporting emits a risk assessment report alongside a sanitized version of the artifact for diffing.
Frameworks & tools
Works with any text-based artifact including source code (Python, JS, Go, etc.), configuration files (YAML, JSON, ENV), and system prompts.
Why this beats prompting it yourself
Manual redaction is prone to oversight, especially with indirect identifiers that don't look like secrets. This skill enforces strict consistency across multiple files, preventing the common mistake of using different placeholders for the same client entity.
Use cases
- Preparing internal tools for open-source release on GitHub.
- Sanitizing configuration logs for technical blog posts or tutorials.
- Cleaning up portfolio projects for job applications or public gists.
- Scrubbing client-specific identifiers from marketplace templates.
Known limitations
Does not perform automated secret rotation at the provider level. Limited to text-based artifacts; binary files or compiled assets are not supported.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
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 today
- Free to download with an account
Needs access to
Acme Corp
Vendor