Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Apx Discord

ASecurity

APX Discord channel — a bot on a community server, an allowlist of rooms with a mode each (always / mention / read), and a sealed turn for everyone. Load BEFORE adding or changing a Discord room, setting the bot up, posting to Discord, or explaining why the bot did or did not answer.

12 stars
0 votes
0 copies
0 views
Added 9/29/2026
ai-agentsgobashawsgitapi

Works with

cliapi

Security Analysis

A100/100

Scanned 9/29/2026

$npx -y skills add agentprojectcontext/apx --skill apx-discord --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Apx Discord?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Apx Discord
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/agentprojectcontext-apx-discord/badge)](https://www.skillsdirectory.com/skills/agentprojectcontext-apx-discord)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: apx-discord
description: APX Discord channel — a bot on a community server, an allowlist of rooms with a mode each (always / mention / read), and a sealed turn for everyone. Load BEFORE adding or changing a Discord room, setting the bot up, posting to Discord, or explaining why the bot did or did not answer.
---

# apx-discord

One Discord bot, run by the daemon plugin `discord`. Config lives at `~/.apx/config.json → discord`; the token is `discord.token` and is never printed or returned by any command.

## Rooms are denied by default

A room that is not listed does not exist: not stored, not indexed, never answered. Each listed room has a mode:

| Mode | Speaks |
|---|---|
| `always` | on every message (a help room) |
| `useful` | when called, and uncalled when a cheap model check (`discord.reply_when`, default: concrete project questions only) says it can help; capped by `limits.gate_checks_per_hour` |
| `mention` | only when called — @mention, a reply to the bot, or one of `discord.names` as a whole word |
| `read` | never; stored and summarised only |

A thread inherits its parent room's mode unless listed itself. Mode changes apply to the next message, no restart. The web panel has the same controls at **Settings → Discord**; saving a new token there reconnects the bot (`POST /api/discord/reconnect`) — from the CLI, a new token or `enabled` needs `apx restart`.

```bash
apx discord status
apx discord channels
apx discord channel set <channel_id> <always|useful|mention|read> [--name <name>]
apx discord channel remove <channel_id>
pbpaste | apx discord set --token-stdin
apx discord set --owner <user_id,…> --names roby
apx discord set --owner-only true    # answer ONLY the owner; everyone else is read, never answered
apx discord set --knowledge-stdin < notes.md   # what the bot can do and answer (≤12k chars)
```

Ids are Discord snowflakes (digits). Never guess one — ask the owner to copy it (Developer Mode → Copy Channel ID).

## Why it did not answer

Decided by code, in this order, before any model runs: room not listed → own message → empty → another bot (stored, never answered) → `read` room → `owner_only` on and the author is not in `discord.owner_ids` (stored, never answered, and never sent to the `useful` gate — with no owner id set, nobody is answered) → `mention` room and not called → per-person cooldown (`limits.user_cooldown_ms`) → per-room hourly cap (`limits.channel_replies_per_hour`). The owner's own account skips the two limits, not the mode. The daemon log says which one applied (`apx daemon logs`).

A burst from one person inside `limits.burst_window_ms` gets ONE answer.

## Every turn is sealed

Owner included: no tools, no private memory, no other channel, no projects. The bot answers from the room's recent messages, its running summary, recall from that room's own index (`discord:<channel_id>` scope) and the owner's notes (`discord.knowledge`, edited in Settings → Discord). Do not tell anyone the bot can look something up in the code or docs — it cannot. If it keeps missing an answer, the fix is the owner's notes.

## Rules and the guardrail

`discord.rules` (Settings → Discord → *Rules for what it says*, ≤2000 chars) is placed LAST in every Discord turn and wins over the notes. Independently of the model, every post goes through `guardDiscordReply` (core/channels/discord/outbox.js): registered secrets masked, local paths replaced, `@everyone`/`@here` defused. Do not tell the owner a rule is "enforced" beyond that — the rules are prompt, the guard is code.

The owner (an id in `discord.owner_ids`, never a display name) is addressed as the owner: by `identity.owner_name`, familiar. Still sealed — the familiarity is tone, not access. Their messages are stored with `meta.owner: true` and the panel draws them on the owner's side.

## Posting

`POST /api/discord/send { channel_id, text }` posts as the bot, only to a listed room, and records it in the ledger. Mentions are disabled at the API level. There is no agent tool for this yet: do not post to a public room unless the owner asked for that exact message.

Attribution

agentprojectcontextagentprojectcontext
View sourceSee grades on GitHubMore from agentprojectcontext →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698461 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →