--- title: Top Communication & Docs Skills for Codex slug: top-communication-docs-skills-for-codex type: category-agent-cut category: communication-docs agent: codex canonical: https://top-agent-skills.com/top-communication-docs-skills-for-codex verified: 6 unverified: 22 updated: 2026-06-29 --- # Top Communication & Docs Skills for Codex For documentation work in Codex CLI, the useful set is documentation and architecture-adr for writing that lives beside the code, system-architecture-diagrams for generating the picture from that writing, and humanizer for anything a customer will read. Notion's skill covers the case where the docs live outside the repo. ## Verified on Codex CLI (6) ### 1. architecture — score 51/100 (official-anthropic) ADR authoring + design review. Trade-offs, consequences, alternatives. - Best for: Choosing between technologies, documenting a design decision, or reviewing a proposal. - Publisher: Anthropic · MIT - Verified on: claude-code, cursor, codex, antigravity ```bash claude plugin install engineering ``` Full entry: https://top-agent-skills.com/skill/architecture-adr (https://top-agent-skills.com/skill/architecture-adr.md) ### 2. documentation — score 51/100 (official-anthropic) Technical writing that engineers will actually read and update. - Best for: Writing the docs nobody on the team has time to write but everyone needs. - Publisher: Anthropic · MIT - Verified on: claude-code, cursor, codex, antigravity ```bash claude plugin install engineering ``` Full entry: https://top-agent-skills.com/skill/documentation (https://top-agent-skills.com/skill/documentation.md) ### 3. incident-response — score 51/100 (official-anthropic) Incident triage + comms + blameless postmortem in one harness. - Best for: Active production incidents and the postmortem afterward. - Publisher: Anthropic · MIT - Verified on: claude-code, cursor, codex, antigravity ```bash claude plugin install engineering ``` Full entry: https://top-agent-skills.com/skill/incident-response (https://top-agent-skills.com/skill/incident-response.md) ### 4. notion-api-skill — score 49/100 (verified-org) Notion API: page/database CRUD, block hierarchies, rich text, schema sync. - Best for: Teams using Notion as a CMS or knowledge base where the agent needs to read and write. - Publisher: Notion · MIT - Verified on: claude-code, claude-api, cursor, codex ```bash npx skills add makenotion/skills ``` Full entry: https://top-agent-skills.com/skill/notion-api-skill (https://top-agent-skills.com/skill/notion-api-skill.md) ### 5. humanizer — score 45/100 (community) Strip the AI tells. Multi-pass editing for content that reads opinionated, not generated. - Best for: Long-form posts where the AI smell would hurt credibility. - Publisher: blader · MIT - Verified on: claude-code, cursor, codex ```bash npx skills add blader/humanizer ``` Full entry: https://top-agent-skills.com/skill/humanizer (https://top-agent-skills.com/skill/humanizer.md) ### 6. system-architecture-diagrams — score 42/100 (community) C4, deployment, sequence-of-events architecture diagrams generated from natural-language descriptions. Full entry: https://top-agent-skills.com/skill/system-architecture-diagrams (https://top-agent-skills.com/skill/system-architecture-diagrams.md) ## In Communication & Docs but NOT verified on Codex CLI (22) These may well work — the SKILL.md format is shared. They are listed as unverified because this site lists an agent only where the install was run or the publisher documents a path for it. - pptx — PowerPoint decks with layouts, charts, speaker notes — programmatic from prose. (https://top-agent-skills.com/skill/pptx) - slack-gif-creator — Animated GIFs sized and optimized for Slack reactions. (https://top-agent-skills.com/skill/slack-gif-creator) - excalidraw-generator — Excalidraw JSON diagrams from prose, with rendering self-checks. (https://top-agent-skills.com/skill/excalidraw-generator) - internal-comms — Internal comms that read consistent. Newsletters, status reports, FAQs, updates. (https://top-agent-skills.com/skill/internal-comms) - doc-coauthoring — Multi-author edits with suggestion mode and conflict resolution. (https://top-agent-skills.com/skill/doc-coauthoring) - frontend-slides — HTML slide decks from prose. Easy to embed, version, and edit later. (https://top-agent-skills.com/skill/frontend-slides) - mermaid-diagrams — Mermaid diagrams from prose with self-validating syntax. Sequence, flowchart, ER, state. (https://top-agent-skills.com/skill/mermaid-diagrams) - linkedin-content-skill — LinkedIn posts that read human. Hook patterns, voice modeling, carousel scripts. (https://top-agent-skills.com/skill/linkedin-content-skill) - knock-notification-best-practices — The multi-channel notification slot nothing else covers: push, in-app and chat alongside email, with copy guidance per channel. No credentials required. (https://top-agent-skills.com/skill/knock-notifications) - education-agent-skills — The pedagogy library: backwards design, rubric generation, formative assessment, learning progressions. (https://top-agent-skills.com/skill/education-agent-skills) - research-co-pilot — The academic writing half nothing else covers: manuscript, grant, peer review, reviewer response. (https://top-agent-skills.com/skill/research-co-pilot) - netsuite-finance-analyst — Oracle's own FP&A analyst: variance decomposition, close readiness, board narratives. (https://top-agent-skills.com/skill/netsuite-finance-analyst) - gtm-skills — 211 GTM plays with the framework named and the output checked. (https://top-agent-skills.com/skill/gtm-skills) - aigtm — The seller’s day: prep the call, run the deal, brief the board. (https://top-agent-skills.com/skill/aigtm) - research-synthesis — Turn raw research into themes, segments, and a prioritized next-steps list. (https://top-agent-skills.com/skill/research-synthesis) - user-research — Research-ops in a skill. Plan the study, run it, synthesize the findings. (https://top-agent-skills.com/skill/user-research) - ux-copy — Microcopy that respects the user. Errors, empty states, CTAs, onboarding. (https://top-agent-skills.com/skill/ux-copy) - brand-review — Brand-voice QA before content ships. Severity-ranked findings with rewrites. (https://top-agent-skills.com/skill/brand-review) - memory-management — Workplace memory that actually understands your team's shorthand. (https://top-agent-skills.com/skill/memory-management) - task-management — Lightweight task tracking in a single markdown file. (https://top-agent-skills.com/skill/task-management) - consolidate-memory — Memory hygiene. Merge, fix, prune. (https://top-agent-skills.com/skill/consolidate-memory) - init — Bootstrap a CLAUDE.md so future sessions start informed. (https://top-agent-skills.com/skill/init-skill) ## Common question **Can Codex keep a docs site in sync with code changes using these?** Partly — the skills supply the conventions and structure, so the agent knows what a good ADR or reference page looks like. Keeping them in sync is a workflow problem, not a skill problem: trigger the docs skill as part of the same task that changes the code, or the docs drift exactly as they always have. --- _Top Agent Skills. Machine-readable twin of https://top-agent-skills.com/top-communication-docs-skills-for-codex._