debug

A structured debugging workflow is a repeatable process for finding and fixing software errors, with logs that record failures and session results.

In plain words
What is it for?
Use it to check linting, type checks, and tests, inspect failure logs, report next steps, and update an error record.
Why use it?
It keeps debugging from becoming a scattered series of guesses and preserves what went wrong for future work.

Command for Claude Code

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 commands/marcusgoll/spec-flow/debug
Clone the repo
git clone --depth 1 https://github.com/marcusgoll/Spec-Flow

Made for: Claude Code.

Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,143 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 $0.00022 $0.04143
Opus 5 $0.00011 $0.02072
Sonnet 5 $0.00004 $0.00829
Haiku 4.5 $0.00002 $0.00414

Measured 2d ago against content hash 2832d12e8f5a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 2d 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.

.claude/commands/phases/debug.md · 508 lines

How it starts

The opening of the file, as written. The whole thing — 508 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/debug — Systematic Error Resolution

Current Branch: !git branch --show-current 2>/dev/null || echo "none"

Feature Slug: !git branch --show-current 2>/dev/null | sed 's/^feature\///' || echo "unknown"

Recent Debug Sessions: !ls -1t specs/*/debug-session.json 2>/dev/null | head -3 || echo "none"

Recent Errors: !tail -20 specs/*/error-log.md 2>/dev/null | grep "^###" | head -3 || echo "none"

Git Status: !git status --short 2>/dev/null || echo "clean"

Debug Session Artifacts (after script execution):

  • @specs/*/debug-session.json
  • @specs//debug/run-XXXXXX/.log
  • @specs/*/error-log.md

Execute debugging workflow by:

  1. Running centralized debug script with arguments: $ARGUMENTS
  2. Analyzing verification status (lint, types, tests)
  3. Reading log files if verification failed
  4. Presenting results with actionable next steps
  5. Updating error knowledge base (error-log.md)

Modes:

  • Manual: User specifies error type and component (non-interactive by default)
  • Structured: Auto-invoked by /optimize with issue metadata (--from-optimize)

Philosophy: Every error is a learning opportunity. Document failures, symptoms, root causes, and fixes. The error-log.md becomes the project's debugging knowledge base.

Workflow position: implement → debug → optimize → preview → ship

Anti-Hallucination Rules

CRITICAL: Follow these rules to prevent fabricating debug results.

  1. Never claim script success without reading artifacts

    • Always Read debug-session.json to verify status
    • Quote actual verification result: "passed" or "failed"
    • Cite log files: "See specs/my-feature/debug/run-abc123/pytest.log:45"
  2. Always read log files if verification failed

    • Don't guess at errors - read actual log content
    • Quote verbatim error messages with file:line references
    • Show stack traces or lint errors from logs

Read the full file on GitHub · 508 lines

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. 2d ago First seen · 508 lines · 22 tokens per session scan A 2832d12e8f5a

Subscribe to this mod's changes

debug is a command published in the GitHub repository marcusgoll/Spec-Flow (91 stars, last pushed 4mo ago), licensed MIT. It adds 22 tokens to every session and 4,143 once invoked, about $0.0001 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 2026-08-30.