AICOM – AI Commander

Full Audit Trail

Every request is logged: timestamp, remote IP, API key label, tool used, parameters, result, and response time. Logs are grouped into named sessions — when an agent opens a session, all its actions are recorded together so you can review, replay, or undo an entire workflow at once. The Audit Logs page includes a session activity chart and a direct Restore button for each session.

Built for Reliable Automation

AICOM speaks strict MCP JSON-RPC: every successful call returns a content field for full client compatibility, protocol errors use standard integer codes, and errors from a tool that actually ran are reported as isError so your agent can see and react to them. Pass an idempotency_key on any write call and a retried request (dropped connection, flaky client) returns the original result instead of repeating the action — no duplicate posts from an accidental double-send. Responses are hardened against corrupted output, so your agent always gets valid JSON back.

Works Well With Small & Local Models

Not every AI agent is Claude or GPT-4 class. AICOM is built to stay usable even for smaller, local, or context-constrained models:

• tools/search — keyword search over tool names and descriptions, tuned for models that don’t know exact tool names: handles full-sentence queries (“how do I add a photo to my page”), common synonyms (shop → WooCommerce, translate → Polylang/WPML), typo tolerance, and plurals.
• tools/describe — fetch one tool’s full parameter schema by exact name, on demand — pairs with tools/search so a model never has to load the entire tool catalog just to find what it needs.
• Compact Tool List (per API key) — an optional setting that strips parameter descriptions from tools/list to cut context size by roughly a third, while always keeping every parameter’s name, type, and whether it’s required — a model can still build a correct call without the extra explanatory text. Off by default; capable models see the full list as before.
• Scoped discovery — tools/list only ever shows the tools a key can actually call. A key without WooCommerce access never sees WooCommerce tools, so a smaller model isn’t wading through options it’s not allowed to use anyway.

Reusable Skills

Let your agent turn a workflow it just performed into a saved, reusable recipe. After finishing a session, AICOM can suggest saving what was done — the exact sequence of tool calls, rules, and expected inputs — as a named Skill . Next time a similar task comes up, the agent fetches the Skill and follows it, instead of re-figuring out the same multi-step process from scratch. Skills are versioned (every edit keeps a full revision history), matched by similarity before creating a duplicate, and scoped like any other capability — a key needs manage.skills to create or edit one.

Accessibility Audits — New in v3.2.0

AICOM now includes a dedicated Accessibility module so your AI agent can audit and fix WCAG issues across your entire site — no external tools or services required:

• Site report — instantly see how many images are missing alt text across your entire media library, with a ranked preview of the top offenders and a percentage score.
• Post audit — scan any post or page for heading hierarchy errors (missing H1, skipped heading levels), images without alt text, and links with non-descriptive anchor text (“click here”, “read more”). Each issue is rated by severity and the post receives an overall accessibility score from 0 to 100.
• Fix in place — set alt text on any media library image in one call. Pass an empty string to correctly mark it as decorative ( aria-hidden ). Full dry-run support so you can preview changes before saving.
• Screenshot before & after — combine AICOM’s audit tools with your AI agent’s browser capabilities to capture a visual record of the page before and after remediation.

A typical AI-driven accessibility workflow: run the site report, get the list of problem images, let the agent analyse each image visually and write descriptive alt text, apply the fixes — then run the audit again to confirm the score improved. All in one session, with a full audit trail.

Supported Modules

• WordPress Core — posts, pages, custom post types, terms, meta, options, menus, plugins
• Media — upload, update, delete media; direct file system access
• Users & Roles — create, update, manage users and role assignments
• Backup — automatic snapshots for posts, terms, and Elementor pages/templates before every write; session-wide restore
• Sessions — named audit sessions with grouped action history
• Accessibility — site-wide alt text audit, post-level WCAG check, alt text fixes
• WooCommerce (optional) — products, categories, settings
• Elementor (optional) — page creation from template, widget inspection, theme builder conditions
• Polylang (optional) — translations, language assignment, string management, and one-call bilingual post pair creation
• WPML (optional) — translation language assignment and translation-group linking for posts and taxonomy terms
• Yoast SEO (optional) — read and write SEO titles, meta descriptions, Open Graph and Twitter card fields; bulk audit across all posts in one session
• SEOPress (optional) — read and write SEO titles, meta descriptions, robots directives, canonical URLs, Open Graph and Twitter card fields; bulk audit across all posts in one session
• Skills (optional) — save, version, and run reusable multi-step workflows
• Clautron (optional) — capability catalog with a machine-readable manifest per capability (version, requirements, scopes, effects, reversibility), install/update/inspect/deactivate lifecycle, blueprint management, primitive verification, event analytics
• ECS (optional) — Ele Custom Skin color schemes, font schemes, custom looks

Who is this for?

• Content teams using Claude, ChatGPT, or any AI assistant who want it to publish and update WordPress content directly
• Agencies managing multiple client sites and wanting AI-assisted workflows with a full paper trail
• Developers building AI-powered WordPress tools or automation pipelines
• Accessibility specialists who want to audit and remediate WCAG issues at scale with AI assistance
• Claude Code users — point AICOM as an MCP server from your terminal and control WordPress alongside your code
• OpenAI Codex users — connect Codex to your site via AICOM’s MCP endpoint and let it manage content as part of your dev workflow
• ChatGPT, Copilot Studio, Dify, n8n users — import the OpenAPI schema URL into any OpenAPI-compatible client; all tools are discovered automatically with Bearer auth
• OpenClaw / Celine / Goose users — native MCP connector, works out of the box

How it works

AICOM exposes a secure HTTP endpoint on your WordPress site. Your AI agent sends structured MCP requests, AICOM authenticates the request, checks scopes and lock state, executes the operation, logs it, and returns a structured response.

AI Agent → AICOM Endpoint → WordPress

API Key Scopes

Each API key is granted specific scopes — you control exactly what each AI agent can and cannot do. Dozens of granular scopes are available, grouped by area:

read.wp, `write.wp.posts`, `delete.wp.posts`, `manage.taxonomies`, `manage.meta`, `manage.media`, `manage.files`, `manage.wordpress.settings`, `manage.plugins`, `manage.backups`, `manage.a11y`, `manage.users`, `manage.roles`, `manage.skills`, plus one `read.*`/`manage.*` pair per optional integration (WooCommerce, Elementor, Polylang, WPML, Yoast, SEOPress, Clautron). Built-in presets (Content Assistant, Tran