paruff

Generate and maintain AI_STANCE.md for any uFawkes* repo. Use when onboarding a repo to the uFawkesAI suite, when reviewing AI policy currency, or when a new AI tool is being adopted. Implements DORA AI Capability 1.

Ensure graphify corpus is current and complete before each agent session. Define minimum AI-accessible documentation per repo. Use at session startup to verify internal context is available to AI tools. Implements DORA AI Capability 3.

15-minute JTBD + acceptance criterion exercise. Use before any spec session. Produces a discovery-brief.md that anchors the entire increment to a real user need and a measurable DORA outcome. Implements DORA AI Capability 6.

Enforce minimum documentation standard across uFawkes* repos. Use when auditing a repo before release, onboarding a new repo to the suite, or implementing roadmap item 0.6 (cross-repo README links). Directly supports DORA AI Capability 3 and the 2023 DORA finding that documentation quality amplifies all technical capabilities.

Create and manage DOJO content for platform engineering education and knowledge sharing.

Maintain AI_STANCE.md as a living document. Use quarterly or when a new AI tool is adopted. Handles review triggers, the three-bucket update process, socialization checklist, and cross-repo consistency. Companion to ai-stance skill.

Full user research methods for uFawkes product decisions. Use when the 15-minute discovery exercise is insufficient — when the riskiest assumption is untested, when user segments disagree, or when a major capability is being designed from scratch. Tier 3: Phase 2.

Compute the four DORA delivery metrics from uFawkesObs (Prometheus + Loki). Use when producing monthly DORA snapshots, validating post-release metric trends, or generating ROI evidence. Requires uFawkesObs running. Implements DORA AI Capabilities 2 and 7.

Architecture Decision Record template and quality checklist with DORA capability linkage. Use when documenting an architectural decision that agents must follow going forward.

Step-by-step OpenTelemetry and uFawkesObs setup: SDK init patterns for TypeScript, Python, Go; DORA metric spans; Grafana dashboard spec. Use when adding observability to a service.

Step-by-step guide to connect a uFawkesAI project to uFawkesPipe and fawkes platform: Dockerfile, ArgoCD manifest, DORA deployment spans. Use when setting up CI/CD for a new service.

Pre-merge security checklist covering secrets, dependencies, auth, data handling, and fawkes suite gates. Use when reviewing a PR for security issues or hardening a change before merge.

TDD patterns and language-specific test examples for TypeScript/Jest, Python/pytest, and Go. Use when writing tests, increasing coverage, or implementing test-driven development.

Add DORA metrics instrumentation to GitHub Actions workflows — job timestamps, deployment tracking, lead time calculation. Load when adding DORA logging to workflows.

GH Actions rules for Fawkes — pin SHA, timeout-minutes, path filtering. Load when editing workflows.

K8s manifest rules for Fawkes — labels, limits, security context. Load when editing platform/ or charts/.

Code quality standards for uFawkes.dev — Jekyll/Liquid linting rules, CSS validation, HTML structure checks, and review thresholds. Used by Review and Build agents.

Full lifecycle for uFawkes components — anatomy definition, implementation spec, styling rules, and Jekyll/Liquid integration patterns. For UI and Build agents.

Handoff protocol, dependency sequencing, and parallel task rules for the uFawkes agent team. Ensures agents don't conflict, block each other, or duplicate work.

Prioritization framework, sprint sequencing, and success criteria for uFawkes.dev work. Used by Planning agent to sequence issues across PRs and agents.

Three-phase accessibility workflow for uFawkes.dev — audit (find issues), remediate (fix specifications), verify (confirm compliance). Covers WCAG AA, Jekyll/HTML patterns, and vanilla CSS focus states.

Content strategy for uFawkes.dev — narrative structure, page-level content hierarchy, tone guidelines, CTA copy patterns, and trust signal rules.

Process for evolving the Fawkes Design System (FDS) — token definition, pattern library maintenance, compliance checking, and documentation. Currently in bootstrap phase.

Hard constraints for uFawkes.dev — Jekyll stack, GitHub Pages limits, Liquid safety rules, CSS conventions, front matter patterns, and commit format. Load this skill before any implementation task.

DORA metric queries for uFawkesObs: PromQL expressions for deployment frequency, lead time, change failure rate, and FDRT sourced from deployment spans processed by the OTel Collector. Load when writing DORA dashboards or interpreting DORA metric data in Prometheus.

Complete uFawkesObs stack reference: service versions, port map, config file locations, Docker Compose operations, and signal flow. Load when configuring, troubleshooting, or extending the uFawkesObs stack.

OpenTelemetry Collector configuration rules for uFawkesObs. Covers pipeline anatomy, common wiring errors, and the specific constraints of the Docker Compose deployment.

Prometheus alert rule patterns and Alertmanager routing config for uFawkesObs. Load when writing new alert rules, tuning thresholds, or configuring Alertmanager receivers.

Grafana Alloy River configuration syntax reference for uFawkesObs. Covers the River DSL, Docker container log discovery, label manipulation, and hot-reload compatibility constraints.

Grafana dashboard JSON patterns, PromQL/LogQL/TraceQL query templates, panel types, and provisioning conventions for uFawkesObs dashboards/. Load when creating or modifying Grafana dashboards.

Grafana dashboard JSON and provisioning configuration rules for uFawkesObs. Covers UID-based datasource references, provisioning file structure, and dashboard JSON conventions.

Shell/Bash and Makefile toolchain: shellcheck, shfmt, make targets, bash 3.2 compatibility rules. Use when writing or reviewing shell scripts, Makefiles, or bash automation in the fawkes scripts/ directory.

Use this skill whenever writing content for a specific stage audience, writing marketing copy, checking module or landing page copy for jargon or inappropriate framing, or translating internal framework language into audience-appropriate vocabulary. Trigger when: writing anything for Magenta, Red, Amber/Mythic, Rational/Orange, Pluralistic/Green, or Integral/Teal audiences — whether module content, marketing copy, QuickStart paths, nav labels, or email sequences. Always load before writing any learner-facing or audience-facing copy. Also load references/vocabulary-bridge-full.md (MKT-07) for worked examples.

Use this skill whenever modifying, creating, or debugging Docusaurus v3 configuration files, sidebar configuration, MDX components, frontmatter, plugin setup, GitHub Pages deployment, or fork-specific configuration for the Integral Education platform. Trigger when: changing docusaurus.config.js, sidebars.js, or any file in src/; installing or configuring plugins; troubleshooting build errors; scaffolding a new fork's Docusaurus config; integrating React components into MDX module files; or fixing heading hierarchy across modules. Load before any file edit in the Docusaurus layer.

Use this skill whenever rating citations, building the Evidence and Citations table (Section 12), evaluating whether a claim requires a citation, or checking whether a module's evidence base is sufficient for publication. Trigger when: any module section makes an empirical claim; when building or auditing Section 12; when the Audit agent flags citation issues; or when the Content Authoring agent is unsure whether a source qualifies as Tier A, B, or C. Also trigger when a module cites findings known to have replication concerns.

Use this skill before scaffolding any fork, before opening a fork PR, or whenever verifying that pre-fork gate conditions are met. Trigger when: the Fork Scaffolding agent is about to create a fork, when checking whether a fork is ready to launch, or when any agent needs to verify which pre-fork issues remain open. Run this skill first and get a documented pass/fail before any fork scaffolding begins. A partial fork creates more remediation work than waiting.

Use this skill whenever building interactive learning components, adaptive assessments, progress tracking, spaced retrieval scheduling, somatic practice timers, partner prompts, or any component that improves the learner's experience beyond static content. Trigger when: creating React components for use in MDX modules, building the competency map, implementing the transition assessment, adding a practice timer or readiness check, implementing the spaced retrieval scheduler, or designing any interactive layer of the platform. Load alongside docusaurus-conventions whenever building components.

Use this skill whenever writing, extending, or auditing an Integral Education learning module. This is the master protocol for the 13-section gold standard every module must meet. Trigger when: creating a new module from scratch, extending an incomplete module, auditing a module against the rubric, writing any individual section (AQAL mapping, practice, Anki cards, retrieval schedule, safety note, etc.), or checking whether a draft meets the platform standard. Always load before touching any module markdown file. Also load references/stage-content-guides.md for the module's target stage before writing.

Use this skill whenever classifying content safety tier, reviewing a module for safety compliance, writing or checking a Safety Note (Section 13), deciding whether clinician review is required, or evaluating whether any practice, prompt, or exercise crosses from Tier 1 into Tier 2. Trigger when: any module contains somatic instructions, shadow work, identity disruption risk, trauma-adjacent content, interpersonal vulnerability, or guided imagery. Also trigger when the Safety Review agent is evaluating a draft, when the Content Authoring agent needs to classify a practice before writing it, or when auditing an existing module's stated Tier.