Skip to content
Skills Library
AI & agents · Tool integration

plugin-installer

/plugin-installer

Install validated Codex plugins from trusted sources with quarantine validation, provenance, and rollback. Use when distribution and installation are the primary goals.

Licence
Apache-2.0
Last repository change
10.10.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
13
Scripts
3
HTML comments
0
URLs in SKILL.md
3
Deprecation

No signals detected. Repository changed on 10.10.2026, without references to retired models or APIs.

Licence

Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.

What to check before installing
  • 3 scripts in this package. Review them before installing, as Claude may execute them.

Scripts: scripts/github_utils.py, scripts/install-plugin-from-github.py, tests/test_wrappers.py

What it does and when to use it

Install validated Codex plugins from trusted sources with quarantine validation, provenance, and rollback. Use when distribution and installation are the primary goals.

Original author description

Install validated Codex plugins from trusted sources with quarantine validation, provenance, and rollback. Use when distribution and installation are the primary goals.

How it works

  1. Philosophy
  2. When to Use
  3. Inputs
  4. Outputs
  5. Execution Boundaries
  6. Workflow
  7. Validation
  8. Anti-Patterns
  9. Constraints
  10. Failure Mode
Original SKILL.md headings · 88 lines, 421 words
SKILL.md preview
---
name: plugin-installer
description: Install validated Codex plugins from trusted sources with quarantine validation, provenance, and rollback. Use when distribution and installation are the primary goals.
metadata:
  short-description: Install validated plugins with provenance and rollback safety
  skill-type: infrastructure_ops
---

# Plugin Installer

## Philosophy

- Install only with provenance and rollback evidence.

## When to Use

Use for downstream plugin installation and visibility recovery after build hardening.

Route elsewhere:
- plugin creation -> `[[plugin-creator]]`
- conversion/hardening -> `[[plugin-builder]]`

## Inputs

- source repo/url and plugin path
- destination root
- pinned ref and trust policy
- validation level (`strict|compat`)

## Outputs

Return: `schema_version`, `installed_plugin`, `install_path`, `validation`, `artifacts`, optional `blocked_by`.

## Execution Boundaries

Classify install work as an external or repo write before acting. Use the OpenAI-style plugin design contract to separate preview/quarantine validation from install, projection refresh, rollback, and marketplace or user-level mutation.

Plugin Installer owns provenance checks, quarantine validation, install evidence, visibility checks, and rollback notes. It does not own plugin scaffolding, plugin hardening, marketplace policy invention, or source-of-truth rewrites outside the requested install destination.

## Workflow

Use the staged install protocol in `references/workflow.md`.

Apply the context-disposition policy: move important still-valid context to references, and intentionally discard stale, duplicated, unsafe, superseded, or low-signal text.

Read when:
- You need full install, provenance, and rollback flow details: [references/workflow.md](./references/workflow.md).
- You need confirmation boundaries for write, destructive, open-world, or completion-gating actions: [OpenAI-style plugin design contract](../../../../../Infrastructure/references/openai-style-plugin-design-contract.md).

## Validation

```bash
bash Infrastructure/scripts/validation-and-linting/validate_skill_authoring_family.sh
```

Fail fast: stop at first failed gate and report blocker text.

## Anti-Patterns

- installing from unpinned moving refs without explicit override
- promoting from quarantine before validation
- omitting provenance or rollback artifacts

## Constraints

- redact secrets and auth tokens in install logs
- do not skip trust policy checks by default
- allow network access only for explicit allowlisted source hosts: `https://github.com`, `https://api.github.com`, `https://raw.githubusercontent.com`

## Failure Mode

- Stop when the source is unpinned, provenance is unclear, trust policy is missing, destination ownership is ambiguous, validation fails, or rollback cannot be described.
- Report the exact blocker and the smallest safe repair instead of partially installing, refreshing projections, or claiming visibility.

## Gotchas

- Quarantine validation is read/prep work; install, projection refresh, rollback, and user-level marketplace changes are stronger side-effect classes.
- A GitHub URL without a pinned ref is not provenance.
- Visibility recovery should not rewrite canonical plugin source unless the source path is explicitly part of the request.

## Examples

- "Install this validated plugin from a pinned GitHub ref and prove it is visible."
- "Quarantine this plugin package first, then tell me whether it is safe to install."
- "Recover plugin visibility without changing the canonical plugin source."

## References

- `references/workflow.md`
- `references/contract.yaml`
- `references/evals.yaml`
- `references/task-profile.json`
- `../../../../../Infrastructure/references/openai-style-plugin-design-contract.md`
- `assets/`
Files in this package (13)
  • LICENSE.txt 11 KB
  • SKILL.md 4 KB
  • agents/openai.yaml 1 KB
  • assets/skill-installer-small.svg 1 KB
  • assets/skill-installer.png 13 KB
  • references/contract.yaml 2 KB
  • references/evals.yaml 5 KB
  • references/task-profile.json 1 KB
  • references/workflow.md 1 KB
  • scripts/github_utils.py 2 KB
  • scripts/install-plugin-from-github.py 1 KB
  • scripts/install-plugin-from-github.pyw 50 KB
  • tests/test_wrappers.py 4 KB

Versions

The database has no other skills named plugin-installer.

Identical content appears in 1 other GitHub folders. We show one copy with the highest score.

Database snapshot: 13.11.2026 · package version 4944660e9698 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90Codex CLI Plugins
    This skill should be used when the user asks to "create a Codex plugin", "build a plugin for Codex CLI", "package a Codex skill", "set up a Codex marketplace", "publish a Codex plugin", "configure plugin.json manifest", "add MCP to a Codex plugin", "install a local plugin in…
  • 100skill-installer
    Use when listing or installing Codex skills from curated sources, GitHub repo paths, private repos, or local package locations.
  • 95skill-installer
    Install Codex skills into $CODEX_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).
  • 85agents-market-installer
    Use when a user asks Codex to recommend, install, update, or inspect specialized coding subagents for Claude Code, Codex, OpenCode, or multiple targets in the current repository.
  • 100skill-installer
    Skill installer - at the start of any new project or new kind of task, find and install the best matching agent skills from GitHub before doing the work. Also when the user asks to install, get or find a skill on GitHub.
  • 90skill-installer
    Install Codex skills into $CODEX_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.