cc-skill-continuous-learning
Development skill skill.
Use this Skill: https://skilld.dev/gh/davila7/claude-code-templates/cc-skill-continuous-learning Nothing lands on disk. Nothing to clean up.
Edit a local copy. It keeps the author and licence.
≈19 tokens for metadata: the name and description. ≈14 when used: this file. ≈98 more on demand in 1 file.
Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.
In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈76 characters, or 0.9%.
Your model, settings, and other Skills decide how much text your Agent can read.
The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.
Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.
Development skill skill.
Source: SKILL.md on GitHub
Ships scripts
skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.
skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.
This skill enables the agent to automatically 'learn' from conversations by extracting patterns, workarounds, and debugging techniques into new skill files. While designed for productivity, this creates a mechanism where malicious instructions provided during a chat could be persisted as permanent agent skills via indirect prompt injection.
No alerts
Risk: LOW · No issues
3/3 files flagged
Score: 93/100 · 2 sections analyzed
Signed by skilld at 053feb2. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.
Last checked against GitHub yesterday.
{
"author": "affaan-m",
"version": "1.0"
}