chinthakat
- 18 skills
- 0 followers
- 3 hours ago last updated
- ▌ Root Cause Investigation · chinthakatInvestigate a defect, an outage or a surprising behaviour down to the actual cause — reproduce first, shrink the reproduction, separate correlation from cause, and stop at the real mechanism rather than the first plausible story. Use for any bug report, incident, flaky test, or "this number looks wrong", and before writing a fix.
- ▌ Deploy Safety · chinthakatBuild, audit or repair a deploy path so it cannot ship to the wrong environment or ship the wrong bundle. Use when adding a deploy script or release workflow, when hardcoded environment names appear in scripts, when two tools disagree about which environment is the default, or after anything reached production that was meant for staging.
- ▌ Contract Change · chinthakatChange a shared contract — an API schema, shared types, an event payload, a database column that something else reads — without breaking the consumers that drift from it silently. Use when editing a schema or shared package, when a local mock or stub has diverged from the real service, or when a change works locally and fails against the deployed system.
- ▌ Design Critique · chinthakatReview a proposed design, plan, architecture or RFC in one adversarial pass and produce ranked findings, each with a concrete failure scenario. Use when asked to review, critique, poke holes in or sanity-check a design before it is built, and before committing to anything expensive to reverse.
- ▌ Evidence Layers · chinthakatMeasure a system that keeps its data in more than one place — source files, staging, production, an archive — without letting a number cross between them. Use before writing any audit, inventory, gap analysis, "what's missing" report or coverage claim, and whenever a finding says something does not exist.
- ▌ Interface Craft · chinthakatBuild or review product UI — the screens people use repeatedly to do a job, as opposed to a marketing page. Covers hierarchy, density, forms, tables, feedback timing, destructive actions, consistency and an accessibility floor. Use when building an app screen, a dashboard, a settings page, an admin console or a form, and when reviewing one.
- ▌ UX State Coverage · chinthakatWork out and design every state a screen can be in — first run, empty, loading, partial, error, denied, offline, stale, overflowing, after-action — and write the copy for each. Use when building or reviewing any screen that loads data or accepts input, and when a bug report describes a blank page, a spinner that never stops, or an unhelpful error.
- ▌ Agent Instructions · chinthakatWrite and maintain the repository instruction file an AI coding agent loads every session (CLAUDE.md, AGENTS.md or equivalent), and decide what belongs there versus in an on-demand skill. Use when setting one up, when an agent keeps repeating a mistake the file was supposed to prevent, or when the file has grown past the point of being read.
- ▌ Data Change Script · chinthakatWrite or review a script that modifies real data — a backfill, migration, cleanup, bulk correction or one-off fix. Use whenever a change cannot be expressed as a code deploy and has to be applied to stored rows, and before running any script someone else wrote against real data.
- ▌ Offline First Sync · chinthakatBuild or debug an app that has to keep working with no network — local writes, a queue, and reconciliation on reconnect. Use when adding offline support, when work is lost on a dropped connection, when sync produces duplicates or silently drops records, or when photos and other large files need to survive being captured out of coverage.
- ▌ Parallel Worktrees · chinthakatWork safely when several people or agent sessions share one repository through git worktrees or parallel checkouts. Use when a repo has sibling worktrees, when changes appear that you did not make, when a stash or branch goes missing, or when setting up a repo for concurrent agent sessions.
- ▌ Access Control Proof · chinthakatAdd or audit an authorization gate — tenant scoping, role or group checks, share links, admin-only routes — and prove that it denies. Use whenever a new endpoint, resolver, query, page or link exposes data that belongs to someone, and whenever a review asks whether a gate actually works.
- ▌ Versioned Publishing · chinthakatDesign or change content, templates, forms, quizzes, pricing rules or configuration that is authored, published, and then consumed by sessions already in progress. Use when editing a published item changes what someone mid-way through sees, when "why did this record's score change?" has no answer, or when retiring an item breaks historical records.
- ▌ Solution Architecture · chinthakatChoose an architecture and record why — constraints before options, reversibility as the ranking axis, at least one option that extends what already exists, and an ADR that survives the people who wrote it. Use when a feature needs a shape rather than a file, when picking between services, stores or patterns, or when someone asks "how should we build this?"
- ▌ Mobile Field Readiness · chinthakatTake a mobile app or PWA feature from "works on my laptop" to "works on a phone in the field" — camera, GPS and geofencing, permissions, install and update, poor signal, one-handed use. Use when building or reviewing anything a person uses standing up, outdoors, on their own device.
- ▌ Requirements Elicitation · chinthakatTurn a vague or solution-shaped request into a statement of what is actually wanted — the job behind the ask, who has it, acceptance criteria that can fail, and an explicit non-goals list. Use at the start of any feature, before designing or planning, and whenever a request arrives already phrased as an implementation.
- ▌ Skill Name · chinthakatWhat this skill does, and the situations in which Claude should reach for it. Name the triggers — the kinds of request, the symptoms, the moments in a task — rather than summarising the body. This is the only part Claude sees before deciding to load the skill.
- ▌ Issue Ledger · chinthakatRecord an issue you found but are not fixing now, so that it is still actionable months later, and close one with evidence. Use whenever an audit, review or unrelated task surfaces a problem that is out of scope, and whenever marking a tracked item done.