The 22 skills · Intranet skill · Everything only

AGENT SKILL

secure-intranet: what it makes your agent do, and what it refuses.

secure-intranet is where reports stop being files and become a system: every report in one place, refreshed on cadence, gated by audience, with shared live state the team sees update. The four layers are never welded together, and the wall in front of them is default deny.

The job secure-intranet does

One dashboard is a file, an intranet is a platform, and the difference is structure. This skill separates landing, data, site and publish so a refresh is a rebuild rather than a bespoke editing session. It puts one site behind an authentication wall with audience zones enforced by access rather than by the build, layers tenant isolation three ways because one way is zero, and sets strict headers and a content policy loosened only per feature. A small worker and database tier carries shared live state for the collaborative surfaces, task boards, checklists, threaded comments and presence, with a reproducible pattern for each. A pre launch gate crawls the real deployment before any external user sees it.

What secure-intranet refuses

A skill is worth what it stops as much as what it produces. These are the refusals written into secure-intranet, taken from the rules and limits in its own file.

• ✓Welding the layers. Data embedded in HTML makes every refresh a manual edit and leaves no history.
• ✓Audience zones enforced by the build. The wall decides who sees what, because a build time decision is not access control.
• ✓Anti discovery as a control. An unlinked URL is a courtesy, never the mechanism.
• ✓Launching on a checklist alone. Configuration drifts and policies get edited in dashboards, so the live crawl against the real deployment is the gate.
• ✓An untested backup. Restores are rehearsed before they are needed rather than after.

6 lines from the secure-intranet skill file, verbatim

From skills/secure-intranet/SKILL.md, unedited. Kymira skills are plain text you can read in full before you install them, which is the point of shipping a method rather than a binary.

[1 LANDING]  data/raw/<cadence>/YYYY-MM-DD__<source>.<ext>   drop zone, archived forever
     │  validate: schema + canary checks + reconcile to the file's own totals
[2 DATA]     data/processed/<cadence>/*.json                 tidy, versioned, page-agnostic
     │  build: one generator per page (template + data → static HTML)
[3 SITE]     site/…                                          static pages + manifest-fed status
[4 PUBLISH]  one deploy, audience zones enforced by AUTH, never by the build

The failure modes secure-intranet addresses

No entry in the catalogue of 51 observed failure modes names secure-intranet as the thing that addresses it. The catalogue records failures we have watched happen, so a skill appears there once a failure it covers has actually been observed and written up. Read the catalogue.

The tier secure-intranet ships in

Both paid tiers are a one-time purchase with lifetime updates, and all sales are final: the files are plain text you keep the moment they arrive, so judge the quality on the free skill and the published doctrine first. The pricing section states exactly what each tier contains.

Installing secure-intranet

A skill is a folder with a SKILL.md file in it. Installing one is copying that folder where your agent looks for skills, with no build step and no dependencies.

mkdir -p ~/.claude/skills
cp -R Kymira/skills/secure-intranet ~/.claude/skills/

Or open the unzipped pack in Claude Code and say install Kymira, which installs every skill in your tier, or double-click the installer for your system. Cursor reads the same skill as a rule file: copy secure-intranet.mdc into .cursor/rules/. Skills are read at startup, so start a new session and ask your agent which skills it has available.

Not for

Last updated 2026-08-27 · Dylan, founder · one of 22 packaged skills, each one a method written down, not a product tour.