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 skills/melodic-software/claude-code-plugins/write-for-agentsnpx skills add melodic-software/claude-code-plugins --skill write-for-agentsgit clone --depth 1 https://github.com/melodic-software/claude-code-pluginsWrote 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/melodic-software/claude-code-plugins/write-for-agents)<a href="https://agentmods.dev/skills/melodic-software/claude-code-plugins/write-for-agents"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/write-for-agents.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 | $0.00200 | $0.01853 |
| Opus 5 | $0.00100 | $0.00927 |
| Sonnet 5 | $0.00040 | $0.00371 |
| Haiku 4.5 | $0.00020 | $0.00185 |
Grade A, and why
write-for-agents 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.
How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write For Agents
Why this skill exists
The docs-hygiene siblings are audit-shaped: they find problems in docs that already exist. This
skill is the write-side complement. It fires while the doc is being written, so the problems the
audits catch are not created in the first place. Its scope is any markdown an agent will consume;
the auto-read surfaces (CLAUDE.md scopes, .claude/rules, auto-memory, and their kin) are the
high-value core because their cost recurs every session. Read
reference/agent-doc-surfaces.md when you need to know
whether, when, and how much of a target file the harness actually loads. Write differently for
an always-loaded surface than for an on-demand one.
Budget both loads
Every line you write spends two budgets, and cutting one can overspend the other:
- Context load. Tokens the agent pays, every session for always-loaded surfaces. Governed marketplace-wide by PLUGIN-PHILOSOPHY's Instruction economy: an instruction earns its place with observed-stumble evidence, or it goes.
- Cognitive load. Attention the human maintainer pays. The human is the index of the doc set: they must be able to hold where things live. Ten tiny fragment files can be cheaper for the agent and ruinous for the human; one 500-line file the reverse. When the two budgets conflict, say which one you spent and why.
Write pointers that cover their branches
A pointer is a routing instruction; the reader decides whether to follow it from the pointer text alone, without opening the target.
- Front-load the leading word. Open with the term the reader is matching on ("Deploys: see…", never "See the following doc for information about deploys").
- Cover the branches. State when to follow it AND what the reader gets ("for tracked-changes output specifically, read X"), so both the follow and the skip are informed decisions.
- A pointer that exists only because changes must be mirrored across distant folders can mask a
cohesion problem. Before adding it, consider restructuring so the things that change together
live together, a pointer papering over low cohesion outlives the reorganization that would
have removed it. (Audit-side remediation home:
claude-memory:audit's C5 fix guidance, if that plugin is installed.)
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 126 lines · 200 tokens per session scan A eac3f4b75bc1
write-for-agents is a skill published in the GitHub repository melodic-software/claude-code-plugins (15 stars, last pushed today), licensed MIT. It adds 200 tokens to every session and 1,853 once invoked, about $0.0010 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
parallel-orchestrator
Manage parallel Claude Code workstreams using git worktrees. Use when: splitting large tasks across multiple workers, coordinating parallel development, monitoring worker progress, integrating completed work, analyzing work item documents (code reviews, issue lists). Triggers: parallel, orchestrator, worktrees…
parallel-worker
Execute focused implementation tasks in a parallel workflow. Use when: working on assigned files in a worktree, making checkpoint commits, signaling dependencies or blockers, completing orchestrator-assigned tasks. Triggers: worker, checkpoint, worktree, assigned scope, commit prefix, parallel task.
build-priority-queue
For ordered processing: A search, Dijkstra, event simulation, task scheduling. Efficient min/max extraction with heap-based queue.
catch-expected-errors
For iteration with errors: catch exceptions during exploration, skip invalid cases, continue to next attempt.
compose-small-helpers
For complex behavior: build from tiny functions, chain transformations, make code read like a pipeline of operations.
count-combinations
For probability and counting: permutations, combinations, sample spaces, Monte Carlo simulation, brute-force enumeration, card/dice problems.