Works with the AI Agents you already use
Secret Redactor
Masks passwords, API keys, tokens, JDBC credentials, private keys, e-mails and internal IPs in code, logs and config files before you paste them anywhere.
Free
Secret Redactor
Example session with this skill installed
Before you paste this stack trace and DataSource config into the GitHub issue, redact anything sensitive.
- Read your context and instructions
- Compiled the secret redactor
redactions: 3 (JDBC_CREDS=2, SECRET=1)
url = "jdbc:mysql://REDACTED:JDBC_CREDS:0d12bd83:REDACTED:JDBC_CREDS:eb02e532@db:3306/prod"
spring.datasource.password=REDACTED:SECRET:66674b8a
Connects securely to your tools. The creator never sees your data.
About this skill
Agents quote code all the time: in docs, issues, pull requests and prompts. This skill makes them redact first.
- One standard-library Python script: no dependency, no network access, nothing written to disk.
- Detects private keys, JDBC credentials, well-known API key formats, secrets assigned to password/token/key-like names (in code,
.env,.propertiesand YAML), e-mails and internal IPs. - A secret found once is masked everywhere else it appears, even under a neutral variable name.
- Markers keep the structure readable: the same value gets the same marker within a run, and no lookup table is ever stored.
--policy blockrefuses to output a file that contains anything sensitive.- Honest about its limits: regex-based, with known false negatives and false positives listed in the skill.
The detection patterns in the script are rules, not secrets: the package contains no credential.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Regex-based. Misses secrets built at runtime, values under neutral variable names, unknown key formats, credentials passed as variables, people's names, internal host names and ticket URLs. May over-mask names that merely contain a keyword (e.g. tokenizerName). Values longer than the bounds (4,096 characters for a secret) are not recognised.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 18 days ago
- Free with an Agensi account
- Software
Python 3.8+
Permissions
File Scopes
No network access. Redacted text goes to standard output.