README

Documentation for reusable subagents: isolated assistants that handle side tasks such as research, verification, or audits and return only their final result. It lists example agents and explains how to add project-specific ones.

In plain words
What is it for?
Use it to understand when to delegate research or checks, and how to define agents for tasks such as fact-checking, regression testing, or migration review.
Why use it?
It keeps detailed side investigations out of the main conversation while preserving focused results for the primary task.

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/muratsilahtaroglu/claude-code-starter-kit/readme
Clone the repo
git clone --depth 1 https://github.com/muratsilahtaroglu/claude-code-starter-kit

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 325 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 $0.00000 $0.00325
Opus 5 $0.00000 $0.00162
Sonnet 5 $0.00000 $0.00065
Haiku 4.5 $0.00000 $0.00032

Measured 2d ago against content hash 2c6c80f824bb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

README 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 2d 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/README.md · 19 lines

What it actually says

.claude/agents/ — reusable subagents

Subagents are isolated assistants for side tasks (deep research, audits, verification) that would clutter the main thread with intermediate results you won't reference again. Each runs in its own context window and returns only its final message to the parent — so the raw exploration never enters your main context. Called via the Agent tool (by name); defined here as a markdown file with name / description / optional tools / model frontmatter + a system-prompt body.

Shipped with the kit:

  • researcher — cited prior-art scout (web + local repo); backs rules.md §8 and /keel-research.
  • verifier — adversarial fact-checker (CONFIRMED / REFUTED / UNCERTAIN); backs rules.md §4.
  • auditor — read-only rules-compliance spot-check over a commit range; backs /keel-audit (the SessionStart hook nudges when one is due).

Add project-specific subagents here as the project needs them — e.g. a regression-runner, a migration-reviewer, a log-analyst. The set is meant to grow per project, not stay fixed. Keep each agent's body focused; put "when to use it" in the description (that's what drives auto-selection). Mechanism trade-offs (skill vs subagent vs rule vs hook): docs/steering.md.

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. 2d ago First seen · 19 lines · 0 tokens per session scan A 2c6c80f824bb

Subscribe to this mod's changes

README is an agent published in the GitHub repository muratsilahtaroglu/claude-code-starter-kit (9 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 325 tokens. 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.