Skip to content
Back to skills

skills

ASecurity

Load full instructions for a skill on demand, list available skills, or read the user's own per-skill additions

  • 13 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 3, 2026
ai-agentsrustbashdocumentation

Works with

  • cli

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 3, 2026

npx -y skills add istota-project/istota --skill skills --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of skills?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for skills
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/istota-project-skills/badge)](https://www.skillsdirectory.com/skills/istota-project-skills)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: skills
description: Load full instructions for a skill on demand, list available skills, or read the user's own per-skill additions
always_include: true
cli: true
---
# Skills loader

Some skills are listed in "Available skills (load on demand)" near the tools
section with only a one-line description — their full instructions are not in
this prompt. Load a skill's documentation before you use it:

```bash
istota-skill skills show <name>    # Print the full instructions for <name>
istota-skill skills list           # List every skill you can load (name + description)
```

`show` prints the same markdown documentation that would otherwise be inlined
in this prompt. Run it once for a deferred skill, read the instructions, then
use that skill's own CLI. Do not guess a deferred skill's subcommands — load
its docs first.

If `show` reports the skill is unknown or unavailable, proceed without it (it
may be disabled or restricted for your account).

## Per-skill overlays

The user can add their own instructions to any skill, in a file of their own at
`{BOT_DIR}/config/skills/<skill-name>.md`. Whatever is in it is appended to that
skill's instructions whenever the skill loads, under a heading saying it takes
precedence — so a skill you loaded normally arrives with its overlay already in
it. The exception is a skill that arrives as a *companion* under another one
(the extra bodies below a `---` in a `skills show`): those carry no overlay, so
fetch it with `skills overlay <name>` if you are about to rely on it.

```bash
istota-skill skills overlays          # what is customized, and does each one load
istota-skill skills overlay notes     # print one overlay
```

These two verbs read; nothing here writes. An overlay is the user's own
document — often prose and code blocks, not a bullet list — so when you are
asked to change one, edit the file directly with your file tools, the same as
any other file in their workspace.

After editing one, run `istota-skill skills overlays` and check that its
`binds` is `true`. That is the only thing that decides whether the file reaches
a prompt, and a file can look right and load into nothing: a misspelled skill
name, a skill that is switched off, a file holding nothing but frontmatter, or
one over the 32 KB cap. The `reason` field says which.

Two skills accept no overlay at all: `sensitive_actions` and `untrusted_input`.

Files in this skill

  • __init__.py22.3 KB
  • __main__.py85 B
  • skill.md2.4 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.