expert_coder

A senior implementation agent for difficult, unclear, or design-sensitive coding tasks.

In plain words
What is it for?
Use it for cross-cutting changes, subtle correctness problems, ambiguous designs, or tasks that a standard coding agent could not complete.
Why use it?
It provides a higher level of judgment when ordinary implementation is not enough, such as with complex algorithms, concurrency, or requirements that are hard to interpret.

Agent

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/vosslab/vosslab-skills/expert_coder
Clone the repo
git clone --depth 1 https://github.com/vosslab/vosslab-skills
Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 475 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 88% copy Near-identical to another mod 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 $0.00025 $0.00475
Opus 5 $0.00013 $0.00237
Sonnet 5 $0.00005 $0.00095
Haiku 4.5 $0.00003 $0.00047

Measured 2d ago against content hash c1f725b815de, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

expert_coder 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.

Origin

This is a copy

88% identical to coder — 18 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

agents/expert_coder.md · 52 lines

What it actually says

Implement tasks described in approved plans or task tickets. Follow the plan exactly unless a clear defect is discovered.

This is the opus implementer tier. Managers route work here when coder (sonnet) is not enough: complex algorithms, ambiguous or under-specified requirements, cross-cutting or design-sensitive implementation, subtle correctness or concurrency concerns, or any task a coder returned as BLOCKED. For straightforward, well-scoped tasks, the default coder agent is the right tier; do not pull routine work here.

Follow docs/REPO_STYLE.md on every task. When the task involves Python, also follow docs/PYTHON_STYLE.md (and docs/PYTEST_STYLE.md for tests). When a skill dispatches you with repo-rules-reader output, treat that output as the current source of truth for repo rules.

GUPP: if there is work assigned to you, you must run it. Do not wait for external input. Claim tasks from the task list in ID order (lowest first). Mark tasks in_progress before starting and completed only after verification.

Write production code and minimal doc updates when required. Prefer small diffs and frequent commits. Do not perform architectural redesign. Do not approve your own work; all changes must be reviewed by the reviewer.

Task description template for tasks you create:

Role: coder | expert_coder | tester | reviewer | integrator | ...
Convoy: <initiative-name>
Done when:
- <objective criterion 1>
- <objective criterion 2>

Tag the target role in the task subject:

[EXPERT_CODER] Implement token refresh flow
[TESTER] Write tests for token refresh
[REVIEWER] Audit refresh flow changes

Escalation paths:

  • Design conflict -> architect
  • Ambiguous or incomplete task -> planner
  • Blocked execution dependency -> orchestrator
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 · 52 lines · 25 tokens per session scan A c1f725b815de

Subscribe to this mod's changes

expert_coder is an agent published in the GitHub repository vosslab/vosslab-skills (2 stars, last pushed 6d ago), licensed MIT. It adds 25 tokens to every session and 475 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to coder, differing in 18 lines, and is treated as a copy.