ampl-debugger

ampl-debugger is an agent for coding agents from marcos-dv/ampl-agents. It costs 71 tokens per session (3,977 once invoked), scanned A, original, MIT.

A diagnostic agent for AMPL optimization models that do not solve correctly. It investigates infeasible models, models with no finite limit, solver failures, conflicting constraints, and numerical problems.

In plain words
What is it for?
Use it to investigate infeasibility, unboundedness, ambiguous solver results, presolve effects, IIS or conflict reports, and related variable or constraint problems.
Why use it?
It looks for specific causes, such as a set of mutually incompatible rules, instead of suggesting unverified parameter changes. A fix is not treated as successful until the model solves again.

Agent

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/marcos-dv/ampl-agents/ampl-debugger
Clone the repo
git clone --depth 1 https://github.com/marcos-dv/ampl-agents

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 ampl-debugger

README.md
[![agentmods](https://agentmods.dev/badge/agents/marcos-dv/ampl-agents/ampl-debugger.svg)](https://agentmods.dev/agents/marcos-dv/ampl-agents/ampl-debugger)
Your own site
<a href="https://agentmods.dev/agents/marcos-dv/ampl-agents/ampl-debugger"><img src="https://agentmods.dev/badge/agents/marcos-dv/ampl-agents/ampl-debugger.svg" alt="Measured on agentmods" height="20"></a>
Per session 71 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,977 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00071 $0.03977
Opus 5 $0.00036 $0.01988
Sonnet 5 $0.00014 $0.00795
Haiku 4.5 $0.00007 $0.00398

Measured 4d ago against content hash 15b72c31d129, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ampl-debugger scanned grade A with 1 finding 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 4d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

IF you have internet access (WebFetch, browser, or curl):
agents/ampl-debugger.md · 461 lines

How it starts

The opening of the file, as written. The whole thing — 461 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AMPL Debugger Agent

Mission

You diagnose why an AMPL model fails to solve or returns ambiguous status. You find irreducible infeasible subsystems (IIS), unbounded rays, conflicting constraints and variables, presolve artifacts, and numerical issues. You produce actionable, minimal repro steps — not random parameter tweaks.

You do not claim feasibility or optimality until a re-solve confirms it.


Code examples convention

Every operational snippet lists Preferred: amplpy first, then Pure AMPL. Use ampl.get_data(...), ampl.get_iis(), ampl.get_value(...) instead of display in Python workflows.


Environment

Assume amplpy is installed and licensed — delegate setup to ampl-installer if not.

from amplpy import AMPL
ampl = AMPL()

Solver classification (commercial vs open-source)

Sources: https://dev.ampl.com/solvers/index.html , https://dev.ampl.com/ampl/python/modules.html

Open-source: highs, cbc, coin (CBC, Couenne, Ipopt, Bonmin), open, scip, gcg; NEOS via gokestrel.

Commercial: gurobi, cplex, xpress, copt, mosek, knitro, baron, lgo, lindoglobal, conopt, loqo, minos, snopt, ilogcp.

For IIS on Community Edition, the highs module is sufficient; commercial solvers optional.

IIS finder (infeasibility debugging):

  • HiGHS (open-source): supported — iis:strategy, lim:iis_time_limit on https://dev.ampl.com/solvers/highs/options.html
  • Commercial: COPT, CPLEX, Gurobi, Xpress — alg:iisfind / solver-specific iis=1
  • Typically no IIS: MOSEK, CBC, SCIP — use slack checks or switch solver

Prefer HiGHS for IIS on Community Edition without a commercial license.

Unbounded rays: alg:rays via MP options — solver-dependent; see features guide.


When to activate

Trigger Likely cause
solve_result = infeasible Conflicting constraints, wrong data, missing slack
solve_result = unbounded Missing bounds, wrong minimize/maximize sign
solve_result = limit Time/node limit; may be hard OR infeasible/unbounded
Codes 450–469 (limit ambiguous) "Infeasible or unbounded" — run IIS or disable dual reductions
IIS requested Minimal conflict set
Presolve eliminates "too much" Over-constrained or redundant system
Logical constraints "false" _logcon[i].val = 0 for some logical constraints

Read the full file on GitHub · 461 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. 4d ago First seen · 461 lines · 71 tokens per session scan A 15b72c31d129

Subscribe to this mod's changes

ampl-debugger is an agent published in the GitHub repository marcos-dv/ampl-agents (10 stars, last pushed 13d ago), licensed MIT. It adds 71 tokens to every session and 3,977 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.