debug

A bounded workflow for reproducing, investigating, fixing, and verifying software failures.

In plain words
What is it for?
Use it for bugs, crashes, flaky tests, regressions, and incorrect behaviour; it narrows the failure, records evidence, tests possible causes, and reports either a verified fix or why the issue remains blocked.
Why use it?
It focuses on finding the underlying cause instead of repeatedly applying guesses or patches that may only hide the symptom.

Command

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/oriolshhh/runware-image-mcp/debug
Clone the repo
git clone --depth 1 https://github.com/Oriolshhh/runware-image-mcp
Per session 50 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 678 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 $0.00050 $0.00678
Opus 5 $0.00025 $0.00339
Sonnet 5 $0.00010 $0.00136
Haiku 4.5 $0.00005 $0.00068

Measured yesterday against content hash f4b03a76d07b, 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 yesterday.

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.

.agent/commands/debug.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/debug — Diagnose and fix a failure

Purpose

Find and correct the causal defect rather than patching symptoms or cycling through speculative edits.

Invocation

/debug <symptom, failing command, error, or reproduction>

Accepted input

Observed and expected behavior, logs/errors, reproduction steps or command, environment, suspected area, and recent changes when known.

Prerequisites

A HarnessKit workspace and permission to run the relevant local reproduction and tests. Read the configured maximum with harnesskit loop list.

Procedure

Follow .agent/loops/debug-loop.yml, counting one full reproduce→investigate→hypothesize→experiment/fix→verify pass as one iteration:

  1. Reuse a supplied context capsule; otherwise apply context-discovery once and construct a task-specific capsule before broad exploration.
  2. Record expected/observed behavior, environment, and exact reproduction.
  3. Reproduce without editing. Reduce the case and trace the narrowest failing path.
  4. Maintain an evidence ledger and rank falsifiable hypotheses. Apply model-routing: a stable, localized reproduction may use standard/medium; retain heavy/high for ambiguous, cross-boundary, concurrent, security-sensitive, or repeatedly falsified investigations.
  5. Run the highest-information controlled experiment, changing one variable.
  6. If evidence identifies the root cause, add a regression test and apply the smallest safe fix. Otherwise begin the next iteration with the new evidence.
  7. Verify the original reproduction, regression, adjacent behavior, and gates.
  8. Stop immediately on verified success. At max_iterations, stop and return the ledger, remaining hypotheses, and next discriminating experiment.

Approval points

  • Ask before destructive, networked, credentialed, production, expensive, or externally visible diagnostics.
  • Requirement or public-contract changes return to /spec for approval.

Expected output

A verified minimal fix with regression coverage and exact commands/results, or a bounded blocked report containing iteration count, evidence, rejected/remaining hypotheses, and next experiment.

Read the full file on GitHub · 74 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. yesterday First seen · 74 lines · 50 tokens per session scan A f4b03a76d07b

Subscribe to this mod's changes

debug is a command published in the GitHub repository Oriolshhh/runware-image-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 678 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.