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/indianic/byline/write-seriesgit clone --depth 1 https://github.com/indianic/bylineWrote 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/indianic/byline/write-series)<a href="https://agentmods.dev/commands/indianic/byline/write-series"><img src="https://agentmods.dev/badge/commands/indianic/byline/write-series.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 | $0.00024 | $0.01024 |
| Opus 5 | $0.00012 | $0.00512 |
| Sonnet 5 | $0.00005 | $0.00205 |
| Haiku 4.5 | $0.00002 | $0.00102 |
Grade A, and why
write-series 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write and publish ONE article using the Byline MCP tools. Arguments: $ARGUMENTS
Run these steps in order. The point of this command is that a correct article costs six tool calls, not the eighteen a series run used to take. Do not add calls the steps below do not ask for.
1. Brief — one call
build_writing_brief with the persona, topic, target site, and word_count.
Choose mode deliberately. blog for anything evergreen; news only when the piece
genuinely turns on events of the last 30 days, and then research is mandatory and must
come from exactly one origin. If it is unclear which, ask the user rather than
guessing.
Read the SCORECARD TARGETS block in the returned brief. It prints the exact integers
score_draft will measure this article against — evidence items, attribution markers,
question headings, first-hand moments, sentence-length spread, paragraph-run cap — already
computed for your word count. Hitting them while writing is the whole reason a revision
round is avoidable. Note the look value from the IMAGES block; every image in this
article must use it.
2. Write the draft
Write the full HTML now, to the brief and to those targets. Two things people get wrong:
- The NEVER USE THESE list in the brief is the exact lexicon that is graded. It is printed from the scorer's own array, so there is no second list to guess at.
- Leave exactly one
[[content_image]]placeholder for the in-body image.
3. Images — one call
generate_images with every image the article needs in a single call: a hero, the in-body
image, and any gallery frames. Pass the same look to each, and give each a distinct
slot.
Do not read the generated files back. The result already carries width, height,
format and mime, read from the image bytes. Reading a PNG through a vision model to
confirm its dimensions is the single most wasteful thing this pipeline used to do.
Check every entry: a batch reports ok: false per image rather than failing as a unit. If
one was refused on safety grounds, reword that prompt and re-run only that slot with
generate_image.
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 · 89 lines · 24 tokens per session scan A fc0d2b2a88bc
write-series is a command published in the GitHub repository indianic/byline (2 stars, last pushed 23d ago), licensed MIT. It adds 24 tokens to every session and 1,024 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 2026-08-31.
Other commands, from other repositories
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.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.