Menu
A skill containing bare base64 secrets ending in equals-sign padding, which must not evade the entropy detector.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Extracts a reusable code pattern from the current repo and authors a portable, validated Agent Skill. Use when you want to capture a logging setup, CI/CD pipeline, observability stack, async messaging pattern, frontend component library, feature-flag middleware, or any other repeatable engineering pattern so it can be replayed across other repos. Produces: a standard-conformant SKILL.md, parameterized code templates in assets/templates/, a PARAMETERS.md documenting every placeholder, then runs validate-skill.sh, scan-leaks.sh, and dry-run-replay.sh as a tiered quality gate. Output is ready to publish via references/publishing.md.
Regression fixture for dry-run-replay header-skip drift.
A clean skill containing a git commit SHA, a UUID, and an embedded base64 image asset — none of which are secrets and none must trip the scanner.
Sets up structured logging for a service using a configurable log level and service name. Produces a logger.config file ready for ingestion by your logging pipeline. Works with any service that reads configuration from a flat config file.
A skill that accidentally contains an AWS access key — should be caught by the scanner.
A clean skill with no secrets or sensitive identifiers. Uses only normal configuration values and parameterized placeholders.
| name | b64-padding-leak-skill |
| description | A skill containing bare base64 secrets ending in equals-sign padding, which must not evade the entropy detector. |
| version | 1.0.0 |
These high-entropy base64 secrets end in = and == padding. A naive
prefix-strip would mistake the padding for an assignment separator and gut the
token; the scanner must still flag them.
PLACEHOLDER_REPLACED_AT_RUNTIME_1
PLACEHOLDER_REPLACED_AT_RUNTIME_2