fledgeling-co/fledgeling-plugins

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Fledgeling Claude Code plugins.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Give agent-authored text a real voice: the prose an agent emits inside Claude Code and other harnesses. Seven registers (terminal reply, work report, commit and PR, review comment, written document, SKILL.md and instruction file, subagent brief), a deterministic lint that hard-fails closing flourishes…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Send a Claude Code agent to work in a container on another machine through Anvil, using the single anvil errand verb, and read its refusals correctly when a piece of the path is missing. Covers the whole loop — check the preconditions, start the errand, watch it, stop it — and translates each stable refusal kind (err.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Keep the portfolio manifest /Dev/ARMADA.md truthful after working in any /Dev project. Surgical, single-entry updates: status, features, key-file references, updated stamp, index row, one changelog line. The counterpart maintenance skill to ship-armada — every project's CLAUDE.md points here so the manifest stays fre.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Takes the open PRs of the Atlas/Bella monorepo to a registered draft release on either the over-the-air or the App Store lane, classifying between them from the expo-updates native fingerprint rather than from intent — ios.buildNumber sits inside that fingerprint, so a buildNumber bump alone forces a store release.…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Validate a UI screenshot against what a test expected and against a design mock, returning a pass/fail gate plus severity-ranked findings. Runs a deterministic pre-scan first (is this evidence, did the surface settle, are the two framings even comparable), crops and zooms rather than judging a full-page thumbnail…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Hold a long autonomous run to a finish line using command hooks and a stall watcher this skill creates itself, rather than the built-in /goal. /goal is a prompt Stop hook judged by a small model on transcript text alone, so it grades what the run said rather than what is true, and Claude Code overrides any Stop hook…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Arm recurring or event-driven work as a change-gated watcher this skill creates itself, so a tick fires only when something has actually changed. A scheduled loop re-sends the same unmet condition and the same failing tasks turn after turn, re-billing the session's whole prefix each time: five of twelve measured heavy.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Decide whether to interrupt the user with a question, then ask it so it takes one click to answer. Sweeps the conversation, the repo and earlier agent output for the answer, kills questions whose answer would not change the work, then refers every surviving technical fork to another model family before the user sees…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

A diff review that reports what it could not check. Finds with fourteen named angles that are forbidden from suppressing each other, so two reasons for flagging one line both survive to be judged; verifies with three verdicts, where realistic-but-unproven stays in as PLAUSIBLE rather than being dropped with the noise.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Write anything in Luke Rhodes' voice, with a copywriting craft layer under the marketing route rather than a voice layer on its own. Routes a request to one of six personas (LinkedIn post and long-form article, marketing content, code review, Slack and chat, short-form, ADHD and book audience), draws the base voice…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

macOS app icon creation with a measured fidelity loop: direction catalogue from a 532-icon corpus, three generation engines (hand-authored layered SVG, Arrow vector, corpus-referenced raster), a written audit.html contact sheet, and a multi-scale scoring harness that iterates the SVG master against the winning raster…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Build a brand-new skill from scratch: a broad clarifying interview that pins down trigger, output and done condition before anything is written, paid+free deep research on the skill's domain (read in full, citations verified), the build itself run through skill-creator with every structural choice traced to evidence…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Scaffold a complete, working new project in /Dev from a project idea — the slipway is where new ships are launched. Interviews up front (modules, React Native/Expo vs native SwiftUI for mobile, macOS window vs menu-bar, codename, bundle prefix, GitHub org, design reference, 1Password vault), then scaffold.sh renders.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Build, review or convert a slide deck across three targets from one body of craft: a self-contained fixed-stage HTML presentation, lecturn.deck/1 JSON with a .pptx round-trip, or an investor deck assembled from a bundled library of 200 slide layouts in 27 families and 21 recipes. Carries its own design layer rather…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Decide which model a piece of work goes to, and produce the exact command that sends it there. One routing policy for every skill that hands work outside the session: six task classes mapped to ten lanes across five model families, each pinned to a model and an effort, with the CLI arguments, the environment GLM needs.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Design or review a user-facing visual artifact — a landing page, app screen, dashboard, clickable prototype, native mobile mockup, wireframe, variation set, tweakable panel, motion piece, 3D hero, print document, generated imagery or design system — as an opinionated designer rather than a code generator. Roots every…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Expert design and UX review of rendered web apps, websites, app UI flows, infographics and design artifacts — the last pass before a human looks at AI-built UI. Runs deterministic accessibility, contrast, target-size, performance and layout-integrity gates first (column alignment, shared rails, section gaps, text…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Session-start operating defaults that cut token spend without cutting task success, tuned for Claude Opus 5 agentic coding. Ships a byte-stable 220-token block for the cached system prefix plus the measured evidence behind every clause, including a paired 106-task benchmark showing that caveman-style response…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

dossier-report: turn a research question into one published, uniquely-themed HTML page — a Dossier paid+free deep-research panel read in full, compiled into a claim graph, then designed from scratch around its own subject and published to .fledgeling.app. Ships three readings of the same argument over one claim g.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Build an interactive, self-contained HTML explainer that makes a hard idea genuinely click - a thing you operate, not a document you read. Lets the mechanism pick the shape of the page from eight that recur (Machine, Field, Solid, Ladder, Fork, Trace, Assembly, Reveal) rather than giving every topic the same tabs, and.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Build a recurring multi-item digest or roundup email that survives contact with real inboxes: a tiered layout that absorbs volume instead of capping it, a table-based render that works in Outlook's Word engine and with images blocked, and a deterministic gate carrying every rule back to the evidence behind it. Refuses.

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Conduct many concurrent Claude Code sessions as one portfolio — the layer above ship-armada, for when the work is already spread across a dozen live sessions rather than waiting in one repo's backlog. Builds a roster of every peer session, learns what each owns, and coordinates them over SendMessage: hands out a…

fledgeling-co/fledgeling-plugins

Plugin Claude Code

Add a Gemini-calibrated gemini.md beside an existing skill's SKILL.md, plus the conditional pointer that other models skip, so one skill works on Google's models as well as Claude's. Most skills in an ecosystem are tuned — silently — to whichever model their author ran, and several of their most deliberate decisions.