AI Agent Workflow Collection

Audit Popoto Redis models for relationship gaps, missing fields, naming inconsistencies, and architectural weaknesses. Use when reviewing data model health, checking model integrity, validating Redis models, scanning for data model issues, or reviewing the data layer with the architect.

Audit tools/ directory for structure compliance, test coverage, CLI quality, and documentation completeness. Use when checking tool health, validating tools, reviewing tool quality, or after adding or modifying a tool. Also use when someone says 'check the tools', 'are our tools documented', or 'which tools need work'.

Pre-publish human-signal gate for social media content. Checks that a draft contains specific metrics, real constraints, and genuine opinions before allowing publish. Required by /linkedin and /x-com before any post goes live. Can also be invoked directly on any draft file.

Use when finding bridge events, agent responses, timeouts, or errors in system logs. Triggered by requests to debug system behavior, investigate errors, or check what the agent did. Always filter by project name.

Audit Claude Code assets — skills, commands, subagents, hooks, and MCP servers — against best practices and optionally apply conservative conformance changes. Use when reviewing overall Claude Code hygiene, bringing the system in line with documented standards, or after adding new skills/commands/subagents. Invoke explicitly; do not use for feature changes.

Use when executing a plan document to ship a feature. Triggered by 'build this', 'execute the plan', 'implement the plan', or any request to run/ship a plan.

Use when sending a spoken debrief to a Telegram chat. Collects context, drafts a 30-second executive brief, synthesizes it via TTS, and delivers it as a native voice message. Triggered by 'send a voice debrief', 'speak this update', 'do-debrief', or any request to deliver an audio summary.

Use when deploying merged changes to production across all bridge machines. Triggered by 'deploy to prod', 'ship it', 'push to prod', or 'do-deploy'. Not part of the SDLC pipeline -- merge already handles the local dev environment.

Template for creating a repo-specific /do-deploy skill. Copy this directory to do-deploy/ and customize DEPLOYMENTPROCESS.md and HEALTHCHECKS.md for your repo's production deployment process.

Audit an existing web UI against premium design criteria. Screenshots pages and evaluates visual hierarchy, typography, color, spacing, consistency, and more. Use when the user wants to evaluate design quality, audit a UI, or says 'review this design', 'check the UI', 'audit this page', 'scan the interface', or provides a URL for design feedback.

Use when translating a moodboard (Cosmos, Pinterest, Are.na, image folder) into additive edits to a design system — tokens, components, charter, downstream CSS. Also use when organizing design files to the canonical docs/designs/ structure (design-system.pen, charter.md, gap-audit.md, inspiration/, product/). Triggered by 'apply this moodboard', 'tighten the design system', 'theme pass', 'design system pass', 'organize design files', or a moodboard URL with a design-system ask.

Discover happy paths on a target site using agent-browser, producing structured trace JSON for deterministic test generation.

Use when cascading documentation updates after code changes. Finds every document referencing the changed area and makes targeted surgical updates. Triggered by 'update docs', 'sync the docs', or any request about documentation updates.

Audit how well a named feature is integrated into its host project. Checks for orphan code, dead wiring, missing tests, undocumented entry points, config gaps, and partial connections. Use when checking feature health, validating a feature shipped correctly, reviewing integration quality, or asking 'is this feature actually wired up?'. Also triggered by 'check feature integration', 'is X connected', 'validate feature', 'scan for dead wiring', or 'what's broken about this feature'.

Use when posting a GitHub investigation issue for an unverified finding, potential gap, or anomaly that needs root-cause analysis before any action is taken. Also use when an audit skill surfaces something suspicious, research reveals a possible gap in the current implementation, or an observed behavior is unexpected but unconfirmed. Issues are labeled 'investigation' only — never 'bug' unless the defect is already confirmed.

Use when merging a pull request that has cleared the SDLC pipeline. Runs the deterministic verify-then-merge gate — confirms the PR is OPEN, mergeable, CI-green, REVIEW-approved, and links its tracking issue — then authorizes and squash-merges. Triggered by 'merge this PR', 'do-merge', or automatically by /sdlc at the MERGE stage.

Audit Python classes and OOP code for structural anti-patterns, naming inconsistencies, and data modeling issues. Use when reviewing class design, checking model health, validating object boundaries, or after refactoring. Also triggered by 'check my classes', 'review the data model', 'are there OOP problems', 'scan for design issues', 'lint class structure', 'audit models', 'validate OOP', or 'review object hierarchy'.

Apply a targeted fix to failing tests or review blockers. Use when the user says 'patch this', 'fix the failures', 'fix the blockers', or 'do-patch'. Also called automatically by do-build at test-fail and review-blocker lifecycle steps.

Use when reviewing a plan before build. Spawns parallel war-room critics (Skeptic, Operator, Archaeologist, Adversary, Simplifier, User) plus automated structural checks. Triggered by 'critique this plan', 'review the plan', 'war room', or 'do-plan-critique'.

Use when reviewing a pull request. Analyzes code changes, validates against plan requirements, and captures visual proof via screenshots. Triggered by 'review this PR', 'check the pull request', 'do a PR review', or a PR URL.

Use when creating a presentation about a feature, concept, or system. Researches the topic, structures content for accessibility, generates diagrams, and exports polished Marp slides styled to the repo's design system. Triggered by 'make a presentation', 'create slides', 'do-presentation', or 'explain this as a deck'.

Use when supervising a full SDLC pipeline run in a local Claude Code session. Loops the /sdlc router — dispatching each stage to a subagent on the stage-appropriate model (opus/sonnet) — until merge, a blocking guard, or the iteration cap. Triggered by 'do-sdlc', 'run the full pipeline', 'ship this issue end to end', 'supervise the sdlc'.

Audit all Claude Code skills for compliance with canonical template standards. Use when checking skill quality, validating skill structure, linting SKILL.md files, verifying frontmatter, or scanning for skill issues. Runs deterministic validation rules and best practices sync against latest Anthropic docs by default.

Use when running the test suite. Parses arguments, dispatches test runners (potentially in parallel), and aggregates results. Triggered by 'run tests', 'test this', or any request about testing.