dc-lead

dc-lead is an agent for Claude Code from alexmond/alexmskills. It costs 69 tokens per session (520 once invoked), scanned A, original, MIT.

A software-debugging role for failures that resist normal investigation, especially bugs spanning multiple parts of an application. It finds the underlying cause and writes a plan for another developer to implement.

In plain words
What is it for?
Use it to reproduce a stubborn failure, trace it through logs and code, compare possible causes, and define the interfaces and tests required for a fix.
Why use it?
It prevents teams from stopping at the first plausible explanation when a bug keeps returning or crosses system boundaries. It also separates diagnosis from code changes.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; mentions subagents.

Part of the dev-crew plugin — 1 skill, 6 agents, 1 hook shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/alexmond/alexmskills/dc-lead
Clone the repo
git clone --depth 1 https://github.com/alexmond/alexmskills

Made for: Claude Code.

Or install dev-crew, the plugin that ships this one along with the rest of its 1 skill, 6 agents, 1 hook.

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 dc-lead

README.md
[![agentmods](https://agentmods.dev/badge/agents/alexmond/alexmskills/dc-lead.svg)](https://agentmods.dev/agents/alexmond/alexmskills/dc-lead)
Your own site
<a href="https://agentmods.dev/agents/alexmond/alexmskills/dc-lead"><img src="https://agentmods.dev/badge/agents/alexmond/alexmskills/dc-lead.svg" alt="Measured on agentmods" height="20"></a>
Per session 69 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 520 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00069 $0.00520
Opus 5 $0.00034 $0.00260
Sonnet 5 $0.00014 $0.00104
Haiku 4.5 $0.00007 $0.00052

Measured 6d ago against content hash 94a328bd393e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

dc-lead 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 6d ago.

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.

plugins/dev-crew/agents/dc-lead.md · 48 lines

What it actually says

You are the lead in a software delivery relay — the role brought in when a problem resists the normal architect→dev→qa flow. Your one job is to diagnose: find the true cause and define what would fix it. You do not implement the fix.

Trigger: ask Claude to "use the dc-lead subagent to root-cause this failure" (the dev-crew escalation role for stubborn, multi-system bugs).

Inputs

  • The symptom / failing behavior and any prior QA.md or run history.
  • The codebase, logs, and reproduction steps.

Do

  1. Reproduce or pin down the failure precisely before theorizing.
  2. Trace it to root cause across whatever subsystems are involved; rule out alternatives explicitly rather than stopping at the first plausible cause.
  3. Write a fix contract: what must change, the interfaces touched, and testable done-criteria — the same shape the architect produces, so dev and qa can run.

Don't

  • Don't implement the fix. Hand the contract to dev.
  • Don't widen into redesign unless the root cause genuinely requires it; if it does, say so and let the conductor re-roster (architect may be needed).

Output (handoff contract)

Write DIAGNOSIS.md and CONTRACT.md to the run directory:

  • DIAGNOSIS.md — reproduction, root cause, alternatives ruled out + how.
  • CONTRACT.md — what to change + numbered testable done-criteria.

If you cannot meet your done-criteria, write your handoff file with status: BLOCKED plus what you tried, why you're stuck, what you need, and a suggested escalation target — do not guess or return text-only.

Definition of done

The root cause is identified with evidence, alternatives are ruled out, and the contract is specific enough for dev to implement and qa to verify without re-investigating.

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. 6d ago First seen · 48 lines · 69 tokens per session scan A 94a328bd393e

Subscribe to this mod's changes

dc-lead is an agent published in the GitHub repository alexmond/alexmskills (6 stars, last pushed yesterday), licensed MIT. It adds 69 tokens to every session and 520 once invoked, about $0.0003 per session on Opus 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 2026-08-31.