Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add OutlineDriven/outline-driven-development --skill create-plugin-scaffoldgit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/outlinedriven/outline-driven-development/create-plugin-scaffold)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/create-plugin-scaffold"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/create-plugin-scaffold.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.00996 |
| Opus 5 | $0.00017 | $0.00498 |
| Sonnet 5 | $0.00007 | $0.00199 |
| Haiku 4.5 | $0.00003 | $0.00100 |
Grade A, and why
create-plugin-scaffold scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- create-plugin-scaffold — 100% identical, 4 lines differ
How it starts
The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create plugin scaffold
Contract
| Field | Bound contract |
|---|---|
| Trigger | The user asks to create an agent plugin or marketplace package |
| Authority | Reversible local: writes only files and directories under a new plugin tree; rollback is deleting the tree. No remote mutation. No VCS mutation. |
| Side effect | Creates a plugin tree on the local filesystem. Scope is bounded to the named plugin directory before any file is written |
| Done | A valid plugin scaffold tree exists and a validation report confirms manifest fields, entry points, and directory layout |
Inputs
- Plugin name (required): the directory name and manifest identifier for the new plugin.
- Target root directory (required): the filesystem path under which the plugin tree is created.
- Target ecosystem (optional): the host agent platform the scaffold targets. Determines the manifest filename, schema, and registration API. Defaults to Claude Code (
plugin.jsonwith.claude-plugin/schema). - Entry point file name (optional): defaults to
index.ts; supplied when a non-default entry is needed. - Plugin manifest fields (optional): author, version, description, and capability list. Apply defaults to omitted fields.
Procedure
- Resolve the target path and verify write access. Confirm the plugin name and target root are present. Resolve the target root to an absolute path and verify it exists and is writable. Compute the plugin directory as
<target root>/<plugin name>. If it already exists and is non-empty, stop and report the collision rather than overwriting. Done when: the plugin directory path is computed and confirmed not to collide with an existing non-empty directory. - Create the standard directory tree. Create the plugin root, an
agents/subdirectory, and askills/subdirectory. Done when: the plugin root,agents/, andskills/directories exist. - Write the host-specific plugin manifest(s) matching the target ecosystem. For Claude Code: write two manifests kept in parity: a root
plugin.jsoncarrying the agent-plugins.org schema ($schema,name,versiondefault0.1.0,description,authoras an object, and anextensionsblock), and a.claude-plugin/plugin.jsoncarrying the Claude manifest schema ($schema,namematching the root,displayName,version,description,author); the repo's gates require matchingnameand parity between the two, so populate both from the same values. For other ecosystems: write the manifest filename and schema that ecosystem requires. Done when: the manifest(s) are written with all fields populated from supplied values or defaults, conforming to the target ecosystem's schema. - Provide the entry surface the target ecosystem requires. For Claude Code: no entry point is needed; the plugin is declarative, loaded by the host from
skills/andagents/. For other ecosystems: write the entry point file at the plugin root using the supplied or default name, exporting the registration function the host calls to load the plugin. Done when: the entry surface the ecosystem requires is present, or a declarative host is confirmed to need none. - Validate the layout against the manifest. Confirm the manifest parses as valid JSON, every declared entry point file exists, the
agents/andskills/directories are present, and no required manifest field is missing. Done when: every validation check passes or every failure is identified.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed · -24 tokens per session e954b3901120
- 3d ago First seen · 45 lines · 58 tokens per session scan A 1a6f6befccc6
create-plugin-scaffold is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed yesterday), licensed Apache-2.0. It adds 34 tokens to every session and 996 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
audit-project
Run an iterative multi-agent code audit until critical and high findings are resolved. Use when the user says "audit my code", "find all the bugs", "deep code audit", "iterative review", or "review until clean".
duet
Use when the user invokes /duet, says "pair with me", or faces aesthetic, architectural, or irreversible decisions.
goal-prompt-drafting
Use when asked to draft copy-ready /goal objectives for long-running agents; returns one normalized one-line objective with measurable end state, grounded proof, easy-out invariants, a stop clause, and a Missing list. Not for source or remote-system changes.
handoff-prompt
Use when the user asks for a handoff, delegation, or clipboard-ready prompt for another agent: a standalone path-free prompt copied to the clipboard, confirmed by title. Not for session-snapshot briefs — use handoff; never remote, credential, publish, deploy, or irreversible.
publish-branch
Use when asked to commit and push to the currently checked-out branch, including the default branch. Not for creating branches, PRs, force pushes, or pushing any other branch.
drill
Use when a concept needs practising rather than explaining: run a scaffolded exercise from worked example to independent problem, quiz the learner, run spaced recall over what they cleared, or probe for the gaps blocking what they want next. For explanation, use explain-concept; for an end-to-end build, use capstone.