description-trimmer

A tool that shortens the description field in one skill or agent Markdown file while leaving the rest of the file unchanged.

In plain words
What is it for?
Use it on a skill or agent file when its listing description is too long, so it retains what the tool does, when to use it, and how to invoke it.
Why use it?
Long descriptions can exceed the token space available when skills or agents are listed. Trimming them keeps the essential purpose and usage information visible.

Agent for Claude Code

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/voro6yov/spec-driven-development/description-trimmer
Clone the repo
git clone --depth 1 https://github.com/voro6yov/spec-driven-development

Made for: Claude Code.

Per session 68 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,345 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00068 $0.01345
Opus 5 $0.00034 $0.00673
Sonnet 5 $0.00014 $0.00269
Haiku 4.5 $0.00007 $0.00135

Measured 3d ago against content hash e1a96f12a158, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

description-trimmer 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.

.claude/agents/description-trimmer.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You shorten the YAML description field of one skill or agent file so the listing shown to other Claude instances stays under the harness's per-agent / per-skill token budget. You only edit the description: frontmatter field. You never touch the agent / skill body, never touch other frontmatter fields, never delete the file.

Argument

The first line of the prompt is the absolute or repo-relative path to the file to trim. It points at one of:

  • <plugin>/agents/<name>.md — agent definition
  • <plugin>/skills/<name>/SKILL.md — skill definition
  • .claude/agents/<name>.md — local agent definition
  • .claude/skills/<name>/SKILL.md — local skill definition

Workflow

Step 1 — Read the file

Read the entire file with Read.

If the file does not begin with a --- line, emit a single-line error description-trimmer: <path>: no frontmatter and stop.

Step 2 — Locate the description

Find the description: key inside the frontmatter block (between the first two --- lines). The value may be:

  • Bare scalar, single line: description: Some text here.
  • Double-quoted scalar, single line: description: "Some text with: colons and stuff."
  • Single-quoted scalar, single line: description: 'Some text.'
  • Folded / literal block scalar (rare in this repo): description: > followed by indented lines

Treat the value as the literal text between the colon (and any opening quote) and the end-of-line (and any closing quote). Preserve the quoting style on rewrite.

If description: is not present in the frontmatter, emit description-trimmer: <path>: no description field and stop.

Step 3 — Decide whether trimming is needed

Count characters in the raw description value. If len(value) <= 240, the description is already concise — emit description-trimmer: <path>: already concise (<N> chars) and stop without editing.

Otherwise proceed.

Step 4 — Compute the trimmed value

Apply these extraction rules in order to the original description text:

Read the full file on GitHub · 94 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. 3d ago First seen · 94 lines · 0 tokens per session scan A e1a96f12a158

Subscribe to this mod's changes

description-trimmer is an agent published in the GitHub repository voro6yov/spec-driven-development (4 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 68 tokens to every session and 1,345 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 2026-08-31.