EmDash is a content management system for creating and managing websites, built with TypeScript and Astro and deployable on Cloudflare or Node.js with SQLite. It is for site owners and developers who want WordPress-like administration, templates, and plugins without PHP or separate hosting.
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 emdash-cms/emdash --skill fixgit clone --depth 1 https://github.com/emdash-cms/emdashWrote 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/emdash-cms/emdash/fix)<a href="https://agentmods.dev/skills/emdash-cms/emdash/fix"><img src="https://agentmods.dev/badge/skills/emdash-cms/emdash/fix/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/emdash-cms/emdash/fix"><img src="https://agentmods.dev/badge/skills/emdash-cms/emdash/fix.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
SkillSpector: 1 finding, up to low
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- low Excessive Agency · line 10 Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.Fix: Limit the skill's scope to its documented purpose. Remove instructions that enable the agent to perform actions outside its stated functionality.
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 | $0.00044 | $0.02165 |
| Opus 5.5 | $0.00018 | $0.00866 |
| Sonnet 5.5 | $0.00009 | $0.00433 |
| Haiku 4.5 | $0.00004 | $0.00216 |
Grade A, and why
fix 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 6d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fix
You are here because a work run established a change it can implement safely. For a bug, verify returned bug, diagnose pinned the cause with at least medium confidence, and diagnose rated the fix mechanical or clear-best-option. Implement the proposed change, prove it works, and leave the candidate verified.
What your output is, and is not. You are not merging and not opening a PR. The trusted publish_candidate tool commits and pushes the durable candidate from a separate publisher sandbox. The model-controlled workspace has no push capability. Publication triggers a preview build; the orchestrator then opens the PR and asks the reporter to try the preview.
Delivery priorities
The working candidate is the deliverable; the regression test is focused evidence. TDD controls ordering, not how much of the run test construction may consume.
- Reuse reproduce's evidence and the repository's existing test infrastructure. One focused case at the lowest layer that proves the same defect is enough.
- Do not add a test configuration, package script, custom harness, or dependency investigation solely to reproduce the bug. Do not inspect
node_modulesunless diagnose identified third-party behavior as the cause. - If the regression test has not converged after three attempts or about ten minutes, the test is at the wrong layer. Switch to a smaller seam. If no meaningful regression test fits existing infrastructure, publish the partial candidate, report the gap, and do not claim the bug fixed.
- Protect the final fifteen minutes for metadata, one verification pass,
publish_candidate, and reporting. Stop optional investigation before that window. - Update the public plan after the first source edit and when moving from editing to verification or publication. Do not leave it on an obsolete test step.
Environment
- Edit with the VFS tools or normal shell commands.
execruns in a credential-free writable container and checkpoints tracked changes back into the durable workspace, including partial output from a failed formatter or generator. - Use local Git when it helps resolve conflicts or inspect history. Local refs and commits have no publication authority. Never push from
exec. - Run tests, lint, typecheck, formatting, and generators with
exec. Run each planned final check once and report its real exit status.
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.
- 6d ago Changed · +8 lines bcf8d700fd16
- 14d ago Changed 1303a0153f2d
- 1mo ago Changed · +1 lines a70f29bd92f1
- 1mo ago First seen · 96 lines · 44 tokens per session scan A c2a7542392e8
fix is a skill published in the GitHub repository emdash-cms/emdash (13,659 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 2,165 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
adversarial-reviewer
Stress-test a code change for concrete correctness defects, unsafe assumptions, and failure modes. Use when the user explicitly asks for an adversarial review, a deep bug hunt, or to tear a change apart. Do not use for ordinary review, style feedback, or general improvement suggestions.
webiny-api-logger-catalog
Name: Logger Import: import { Logger } from "webiny/api/logger" Source: @webiny/api-core/features/logger/index.ts Description: Structured logging with multiple log levels.
creating-plugins
Create EmDash CMS plugins with sandboxed hooks, routes, storage, content and media APIs, MCP tools, and declarative admin UI, or native React and Astro extensions. Use when scaffolding or implementing an EmDash plugin.
emdash-cli
Use the EmDash CLI to inspect and manage an EmDash instance from the command line, including content, schema, media, taxonomies, menus, search, authentication, seeds, migrations, generated types, and whole-site export and import.
wordpress-plugin-to-emdash
Analyze and port WordPress plugin behavior, custom post types, shortcodes, admin workflows, and stored data to current EmDash extension points. Use for WordPress-plugin migrations or when deciding which behavior belongs in an EmDash plugin, site schema, seed, or Astro code. Do not use for visual theme ports without…
building-emdash-site
Build the site-facing parts of an EmDash CMS project on Astro, including schema and seeds, content queries, Portable Text rendering, menus, taxonomies, widgets, and deployment configuration. Use for EmDash site and theme work; use creating-plugins instead when the task is plugin authoring alone.