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 agentmods add commands/mikefluff/skills/paydowngit clone --depth 1 https://github.com/Mikefluff/skillsWrote 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/commands/mikefluff/skills/paydown)<a href="https://agentmods.dev/commands/mikefluff/skills/paydown"><img src="https://agentmods.dev/badge/commands/mikefluff/skills/paydown.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00013 | $0.01525 |
| Opus 5 | $0.00006 | $0.00763 |
| Sonnet 5 | $0.00003 | $0.00305 |
| Haiku 4.5 | $0.00001 | $0.00153 |
Grade A, and why
paydown 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 3d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Paydown + QA sweep
Clear structural debt from scripts/code-quality-baseline.json and fix whatever
else the sweep turns up. Argument (optional): a cluster name from the table
below, or a file path. With no argument, work the table top-down.
$ARGUMENTS
Read first
python3 scripts/check-code-quality.py --report | grep -v '^!' | head -40
python3 scripts/check-code-quality.py # must be green before you start
git log --oneline -8
CONTRIBUTING.md §2b describes the gate. The short version: thresholds are
module ≤400 lines, function ≤50, branch complexity ≤12, parameters ≤5. Known
violations are frozen; anything new is a hard failure. The baseline may only
shrink — --freeze refuses to grow it.
The state as of the last session
114 frozen violations across 48 files. Zero in the publishing layer — that one
was refactored to zero and is the worked example of what "done" looks like
(publishers/rules.py, publishers/_oauth.py, postsource.py,
cli/_publish_view.py, cli/_reel_stitch.py).
| # | Cluster | Count | The move |
|---|---|---|---|
| A | -maker CLI main() × 9 (carousel, cover, flyer, banner, thumbnail, meme, quote, logo, avatar) |
~25 | Start here. They are near-identical: parse → resolve style → build plan → estimate → confirm → batch → print. Extract one skeleton into cli/_maker.py and let each module declare its differences. Nine separate edits would be the wrong answer. |
| B | Other CLI (gif, styles, stylize, proposal, transcribe, upscale, subtitle, bg, mix, _shared) |
~20 | Case by case. cli/styles.py (537 lines) and cli/proposal.py (main() at complexity 41) are the big two. |
| C | providers/*.py — generate() / poll() |
17 | Shared shape: build body → POST → handle error → poll → download. kling, suno, google_video are the worst. A providers/_polling.py helper likely collapses most of it. |
| D | proposal_*.py (kit, parse, render, brand) |
15 | parse() and render_html() both run >100 lines at complexity ~30-40. Real decomposition, no shared skeleton to lean on. |
| E | skills/writer/scripts/lint.py (988 lines) |
6 | Highest risk, do last. It is the base linter every skill depends on, and 31 fixture snapshots pin its behaviour. Split by concern (catalogue / structural / rhythm / report) and lean on bash tests/run.sh after every step. |
| F | runners core (ffmpeg 7, typography 5, styles 5, batch 3, misc 4) |
~24 | Mostly parameters findings — signatures that grew into bags. A frozen dataclass per call is usually the fix, as PostOverrides was. |
| G | scripts/*.py |
7 | Small and safe. Good warm-up. |
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.
- 3d ago First seen · 98 lines · 13 tokens per session scan A 7a299aae8dc1
paydown is a command published in the GitHub repository Mikefluff/skills (18 stars, last pushed 25d ago), licensed MIT. It adds 13 tokens to every session and 1,525 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.
Other commands, from other repositories
clarity-interview
Co-write a piece with clarity, starting from an interview.
clarity-review
Review a draft with clarity, without rewriting it.
clarity-rewrite
Rewrite a draft with clarity.
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.