agentconnect: Skill for Claude Code

.claude/skills/agentconnect-debug/SKILL.md

agentconnect-debug is a skill for Claude Code, Codex from agentconnect-md/agentconnect. It costs 163 tokens per session (2,122 once invoked), scanned A, original, Apache-2.0.

An evidence-based troubleshooting guide for live AgentConnect incidents, such as stuck sessions, failed hooks, unreachable agents, interrupted turns, or daemons that drop work. A daemon is a background service that runs continuously.

In plain words
What is it for?
Use it to establish the correct environment and identifiers, inspect logs and control-plane records, and work through evidence in a fixed order.
Why use it?
It prevents investigations from checking the wrong environment, old code, wrong logs, or a single indirect signal as if it were proof.

Skill for Claude CodeCodex

Written for Claude Code and Codex: installed under .claude/, but also agents/openai.yaml present.

This is agentconnect-md/agentconnect's own configuration. It tells Claude Code and Codex how to work on agentconnect itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything agentconnect configures →

About the project

AgentConnect is an open-source platform that lets teams and multiple AI agents collaborate in shared conversations and work processes across services such as Slack, Telegram, Discord, GitHub, GitLab, and Linear. Teams use it to assign agents roles, have agents call one another, and start work from messages, issues, pull requests, webhooks, or schedules.

agentconnect-md/agentconnect · 1,520 stars · on GitHub · agentconnect.md

Reuse

Borrowing it

Nothing to install: this file belongs to agentconnect-md/agentconnect. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/agentconnect-md/agentconnect/main/.claude/skills/agentconnect-debug/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/agentconnect-md/agentconnect

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for agentconnect-debug

README.md
[![agentmods](https://agentmods.dev/badge/skills/agentconnect-md/agentconnect/agentconnect-debug/github.svg)](https://agentmods.dev/skills/agentconnect-md/agentconnect/agentconnect-debug)
Your own site
<a href="https://agentmods.dev/skills/agentconnect-md/agentconnect/agentconnect-debug"><img src="https://agentmods.dev/badge/skills/agentconnect-md/agentconnect/agentconnect-debug/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agentconnect-debug

Your own site · 80×15
<a href="https://agentmods.dev/skills/agentconnect-md/agentconnect/agentconnect-debug"><img src="https://agentmods.dev/badge/skills/agentconnect-md/agentconnect/agentconnect-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 163 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,122 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00163 $0.02122
Opus 5.5 $0.00065 $0.00849
Sonnet 5.5 $0.00033 $0.00424
Haiku 4.5 $0.00016 $0.00212

Measured 1mo ago against content hash 1088d2f555e9, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

agentconnect-debug scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/cp-query.sh, scripts/daemon-store.cjs), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/agentconnect-debug/SKILL.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debug AgentConnect

Work from evidence, in a fixed order, and say what each conclusion rests on. Most hours lost in past investigations were not spent on hard problems. They went to four mistakes: reading the wrong environment, reading main while the daemon ran an older release, searching the place one happens to know instead of the place the artifact must be, and treating one indirect signal as proof. The order below forces those checks before anything deeper.

0. Establish the coordinates

Take the environment coordinates from local instructions or memory, never from names or guesses: which console and Control Plane is which environment; where each self-hosted daemon runs, with its root, service form, and log source; the cluster and namespace of pool daemons; the local time zone the console renders in; and any read-only restrictions on production. A tool connector or console login bound to one environment tells you nothing about another. If the coordinates are not recorded, ask for the environment and the daemon before reading anything.

Write down, before touching data:

  • the environment and the ids in the report — session, agent, daemon, hook run, check run;
  • the time window in UTC (database timestamps are UTC; the console may render a local zone);
  • which platform the session belongs to (session_meta.platform, hookKind), because that decides the ingress path.

1. Pin the version

Self-hosted daemons lag the release train by one or more release candidates; pool members follow it. Behavior that the current code cannot produce is usually old code, and a Control Plane table is written by daemons of several versions at once, so one odd row proves nothing about main.

Read the writer's version first (daemon: name, agentVersion, lastSeenAt, host), then read the code at that tag:

git show v<agentVersion>:packages/daemon/src/<file>

To test what main does, use an agent placed on the pool, not a self-hosted daemon. When a symptom's timestamp is within a minute of a release, read the deployment's rollout times before reading code: a Control Plane rollout leaves a window where daemons are reconnecting.

Read the full file on GitHub · 80 lines

Files

What ships with it

6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 1mo ago First seen · 80 lines · 163 tokens per session scan A 1088d2f555e9

Subscribe to this mod's changes

agentconnect-debug is a skill published in the GitHub repository agentconnect-md/agentconnect (1,520 stars, last pushed today), licensed Apache-2.0. It adds 163 tokens to every session and 2,122 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-27.

Related

agentconnect-debug compared with skills from other repositories

agentconnect-debug next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
agentconnect-debugagentconnect-md/agentconnectSkill1.5ktodayApache-2.0A163Claude Code, Codex
python-code-qualitymicrosoft/agent-frameworkSkill15kyesterdayMITA40Claude Code, Codex
backtest-diagnoseHKUDS/Vibe-TradingSkill38kyesterdayMITA25Claude Code, Codex
oma-observabilityfirst-fluke/oh-my-agentSkill1.4kyesterdayMITA31Claude Code
node-inspect-debuggermateaix/mateclawSkill1.2kyesterdayApache-2.0A19Claude Code, Codex
issuejeremylongshore/tons-of-skills-marketplaceSkill2.9kyesterdayMITA115Claude Code
browser_cdpmateaix/mateclawSkill1.2kyesterdayApache-2.0A44Claude Code, Codex