heurema

Use when the user wants to set up or update their content profile. Also use when the user says "my blog is about", "content profile", "set up content context", or "who am I writing for". Stores context as .agents/content-context.md — read by all other content-ops skills first.

Use when the user wants content ideas or a backlog. Triggers: "content ideas", "what to write next", "build backlog", "ideas from changelog", "what should I post about", "content calendar".

Use before publishing any article. Triggers: "check article", "preflight", "validate post", "is this ready to publish", "QA my post", "pre-publish check". Run this before distribute or any publish command.

Use when the user wants feedback on marketing copy, headlines, or landing page text. Triggers: "edit landing page", "improve headline", "copy feedback", "review this copy", "is this headline good", "landing page copy review".

Use when the user wants to publish an article to multiple platforms at once. Triggers: "publish article", "distribute to platforms", "send to devto and hashnode", "cross-post", "publish everywhere", "distribute my post".

Use when the user wants to write a welcome sequence or email drip campaign. Triggers: "welcome sequence", "newsletter drip", "onboarding emails", "email series", "welcome email". Note: generates drafts only — no sending, no API.

Use when the user wants to announce a product launch or new feature. Triggers: "HN Show post", "announce feature", "product launch post", "announce on HN", "write launch announcement", "post on HN", "Product Hunt post", "IndieHackers post", "GitHub launch".

Use when the user wants to plan and execute a multi-platform launch. Triggers: "launch plan", "launch sequence", "launch checklist", "multi-platform launch", "coordinate launch", "launch timeline".

Use when the user wants contract-first development — define correctness before coding, implement against a contract, audit with multiple models, and package proof artifacts.

Use when the user wants a contract-first implementation workflow, asks to define correctness before coding, wants multi-stage execution with verification and audit, or needs a proof-style artifact for a code change. Codex should run a CONTRACT -> EXECUTE -> AUDIT -> PACK pipeline.

Run the Rust CAK host adapter gate from an Agent-Skills-compatible wrapper.

Review CAK R&D PRs and research packets for unsupported claims, overclaimed readiness, lifecycle violations, stage mismatches, and missing proof grounding.

Scan and aggregate task checkboxes from Speckit (.specify/) and context-planning (specs/) formats into a unified backlog.yaml, with automatic project context analysis and auto-update capabilities.

Generate daily and weekly planning reports from backlog and carryover state, applying WIP limits and priority rules from BaseContext.yaml, with automatic git commit/push.

Use Pactum's contract-first workflow for non-trivial repository code changes. Use when a code task needs contract approval, a prompt boundary, or a safe plan step. Do not use for quick questions, trivial edits, or when the user asks not to use Pactum.

Daily curated news digest for your domain. One command to set up, zero API keys, works offline after first fetch.

Use when making a meaningful change in the Punk repository. Guides Codex through Punk bounded work discipline: run Research Gate preflight, read work/STATUS.md, follow selected_next, preserve scope, update goal/report/status, and run checks. Do not use for explanation-only questions.

Use for evidence-grounded deep research runs, reference atlases, technical landscape maps, literature reviews, market maps, idea-generation research, and strategic memos. Creates structured run artifacts before final synthesis.