AIRS_Data_Analysis

Create agents that pass agent-review's six gates by construction — role capture, distinct-from-skill check, tool allowlist minimization, draft against gates, dogfood self-review. Use when authoring a new agent, refactoring an existing one, or promoting a Mall agent into the heir's brain.

Audits a candidate agent (.agent.md) against five gates (spec compliance, content quality, scope fit, safety, currency & coherence) plus Gate 6 (tool allowlist minimality). Use when reviewing a new agent draft before commit, evaluating a Mall agent or store agent for adoption, or re-auditing existing agents on a periodic cadence.

Detect, resolve, and manage the Alex_ACT_Memory shared memory bus. Fires on bootstrap, session start (announcements), and feedback writes.

Generate on-brand Alex — ACT Edition SVG banners for documents (READMEs, plans, notes, release artifacts)

Prevent fabricated facts, invented APIs, and citation confabulation at the point of generation. Use when generating factual claims, code examples, API references, library names, configuration values, error messages, or citations — anything where 'sounds plausible' is not the same as 'is real'.

Perform a local brain audit for ACT Edition (and Supervisor) using deterministic QA plus targeted file review, then produce severity-ranked fixes. Pairs with extension-audit on the sibling surface side; the Marketplace surface routes there, not here.

Systematic code review for correctness, security, and growth — not just style enforcement

Scaffold book-writing projects with folder structure, BOOK-PLAN template, character development matrices, and chapter organization patterns. Use when starting a new book project, planning multi-chapter structure, or building character bibles.

Challenge what you think is right — alternative hypotheses, missing data, evidence quality, bias detection, falsifiability, and adversarial review

Adversarial code review with three parallel perspectives — Advocate, Skeptic, Architect — that create productive tension. Use for high-stakes PRs, architectural changes, or when single-pass review would miss issues. Surfaces findings through disagreement, not consensus.

Documentation hygiene — anti-drift rules, count elimination, and living document maintenance

Convert Word documents (.docx) to clean Markdown with image extraction and pandoc cleanup

Apply consistent git practices for branch hygiene, safe commits, and recovery from common mishaps (lost commits, bad merges, accidental pushes). Use when authoring or reviewing a git workflow, recovering broken local state, or sequencing a commit/push that needs explicit user approval before destructive steps.

Greeting-triggered self-check — recognise greetings, check Edition version against the upstream tag, scan AI-Memory announcements, and report inside the greeting reply

Convert HTML documents to clean Markdown via pandoc

Create instructions that pass instruction-review's five gates (plus Gate 6 for always-on) by construction — intent capture, prior-art scan, applyTo calibration, draft against gates, dogfood self-review. Use when authoring a new instruction, refactoring an existing one, or promoting a Mall instruction into the heir's brain.

Audits a candidate instruction (.instructions.md) against five gates (spec compliance, content quality, scope fit, safety, currency & coherence) plus optional Gate 6 (token budget for always-on instructions). Use when reviewing a new instruction draft before commit, evaluating a Mall instruction or store instruction for adoption, or re-auditing existing instructions on a periodic cadence.

Write markdown that passes markdownlint on first attempt — encode the most common rules as muscle memory

Author markdown and Mermaid diagrams that render correctly in VS Code, GitHub, and other Mermaid 10+ consumers — covers diagram-tool selection, mandatory init template, parser pitfalls, and visual design rules. Use when writing technical docs with embedded diagrams, debugging Mermaid render failures, or choosing between Mermaid, Excalidraw, and other diagramming tools.

Render user-supplied markdown safely — marked.js → DOMPurify → Mermaid (order matters; skipping the sanitizer is XSS)

Convert Markdown to RFC 5322 email (.eml) with inline CSS and CID images

Convert Markdown to standalone HTML pages with embedded CSS, images, and Mermaid diagrams

Strip Markdown formatting and produce clean plain text via pandoc

Convert Markdown with Mermaid diagrams and SVG illustrations to professional Word documents

Consolidate session learning into permanent architecture — extract patterns into skills, instructions, prompts, or memory

Step-back protocol — restate, generalise, specialise, invert, ask why, pre-mortem, check stakeholders, and audit framings before solving

Create prompts that pass prompt-review's five gates by construction — intent capture, slash-command naming, single-workflow scope, draft as numbered steps, dogfood self-review. Use when authoring a new prompt, refactoring an existing one, or promoting a Mall prompt into the heir's brain.

Audits a candidate prompt (.prompt.md) against five gates (spec compliance, content quality, scope fit, safety, currency & coherence). Use when reviewing a new prompt draft before commit, evaluating a Mall prompt or store prompt for adoption, or re-auditing existing prompts on a periodic cadence.

Hardens code against vulnerabilities. Use when handling user input, authentication, data storage, or external integrations. OWASP-aware, language-agnostic principles with TypeScript examples — applies to any feature that accepts untrusted data, manages user sessions, or interacts with third-party services.

<≤300 chars, third-person, behavioral, names what + when (trigger phrases)>

Append a one-line entry to docs/ledgers/curation-log.md whenever a brain artifact is added, modified, or removed. Use when committing changes to .github/skills, .github/instructions, .github/prompts, or .github/agents.

Create skills that pass skill-review's five gates by construction — intent capture, prior-art scan, draft against gates, dogfood self-review. Use when authoring a new skill, refactoring an existing one, or adopting a Mall unit into this brain.

Audits a candidate skill (.github/skills/<name>/SKILL.md) against five gates (spec compliance, content quality, scope fit, safety, currency & coherence). Use when reviewing a new skill draft before commit, evaluating a Mall unit or store skill for adoption, or re-auditing existing skills on a periodic cadence. For instructions, prompts, agents — use the matching per-type review skill.

Create stakeholder-friendly project status updates and progress reports

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes. Four-phase root-cause-first method (investigate → pattern-analyze → hypothesize → implement) that beats guess-and-check thrashing.

Mermaid renders blank or garbled — diagnose silent failures in timeline, gitGraph, gantt; default to flowchart for arbitrary text