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.
npx skills add flonat/flonat-research --skill test-iterate-loopgit clone --depth 1 https://github.com/flonat/flonat-researchWrote 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.
[](https://agentmods.dev/skills/flonat/flonat-research/test-iterate-loop)<a href="https://agentmods.dev/skills/flonat/flonat-research/test-iterate-loop"><img src="https://agentmods.dev/badge/skills/flonat/flonat-research/test-iterate-loop.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00055 | $0.02308 |
| Opus 5 | $0.00028 | $0.01154 |
| Sonnet 5 | $0.00011 | $0.00462 |
| Haiku 4.5 | $0.00006 | $0.00231 |
Grade A, and why
test-iterate-loop 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 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.
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.
How it starts
The opening of the file, as written. The whole thing — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Test-Iterate Loop — Autonomous Bug Fix Cycle
Autonomous loop: run tests → root-cause failures → apply minimal fix → retry. Bounded by iteration cap and same-error-repeat detector. Never commits — leaves clean working tree + markdown report. Generic across Python (pytest), R (testthat), Julia (Pkg.test), and HPC pipelines (mock-Avon Docker).
Hard Rules
Existential — block proceed
- Never
git commit,git push, or modify.git/state. All changes go into the working tree only. The user reviews the final clean tree + report and decides what to commit. Enforced via the standard forbid-list persubagent-write-guard.mdif dispatching sub-agents. - Bounded iteration: max 10 iterations OR 3-same-error-repeat — whichever first. Hardcoded ceiling. Past that, stop and summarise. Don't ask "continue?" — exhaustion is the signal.
- Each iteration must change ≥1 file. A no-op iteration (Claude couldn't propose a fix) counts as a same-error-repeat.
- Memory bug detector: flag any iteration that adds a model load, env var change, or version pin without a corresponding test — these are common failure-cause patterns from past HPC sessions.
Format — catch in review
- Every iteration logged to
log/test-iterate/<project>-YYYY-MM-DD-HHMM.mdwith: hypothesis, fix applied, result. - Final report at the same path summarises iterations, terminal state, and remaining failures (if any).
- Use
TaskCreate/TaskUpdatefor live progress tracking — the user can see what iteration is running and why.
When to Use
- A pipeline / library has failing tests and you want them fixed autonomously
- Pre-flight before submitting an HPC job (catch torch/transformers/CUDA mismatches in mock-Avon Docker before the real job queues)
- Refactor cycles: change → run tests → fix breaks → repeat
- Reproducing a failure on a fresh checkout
When NOT to Use
- The test suite itself is broken (fix the tests first, or you'll loop trying to fix code to match wrong tests)
- The failure root cause is external (network, third-party API, hardware) — the active agent cannot fix those
- You want to write tests, not fix them — use
computational-experimentsor direct work - The fix requires research / design decisions, not just code adjustment
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.
- 4d ago First seen · 198 lines · 55 tokens per session scan A c8456fbdb4d7
test-iterate-loop is a skill published in the GitHub repository flonat/flonat-research (132 stars, last pushed 13d ago), licensed MIT. It adds 55 tokens to every session and 2,308 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 2026-09-03.
Other skills, from other repositories
ulw-qa
This is a Hermes-native ultraqa workflow skill.
boundary-probing
Map parameter space, generate extreme values, test at boundaries, detect breakpoints, synthesize validity envelope.
ia-verification-before-completion
Enforces fresh verification evidence before any completion claim. Use when about to claim "tests pass", "bug fixed", "done", "ready to merge", handing off work, or before editing when a request has ambiguous scope.
check-pipeline
Check that a code and its Pipeline/ .md are still compatible — that every symbol, cell number, data file, and data-flow claim in the pipeline doc still matches the current code (drift detection). Use after editing a documented notebook/engine/script, before trusting a pipeline doc, or on request ("does the pipeline…
reality-check
Re-derive a contested or pushed-back claim in isolation from the definitions, to catch sycophantic capitulation. Use when a result is disputed or you suspect you changed an answer only because the user pushed back.
refactor-ops
Safe refactoring patterns - extract, rename, restructure with test-driven methodology and dead code detection. Use for: refactor, refactoring, extract function, extract component, rename, move file, restructure, dead code, unused imports, code smell, duplicate code, long function, god object, feature envy, DRY…