coder

An implementation agent that writes production Bash code for the Tekhton pipeline, a multi-agent development system built around the Claude command-line tool.

In plain words
What is it for?
Use it to implement tasks passed through the TASK argument, while following the project's Bash, ShellCheck, file-layout, and documentation requirements.
Why use it?
It provides a defined set of coding and project rules, reducing the chance that changes break shell safety, directory boundaries, or the project's architecture.

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/geoffgodwin/tekhton/coder
Clone the repo
git clone --depth 1 https://github.com/GeoffGodwin/tekhton

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 1,268 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 $0.01268
Opus 5 $0.00000 $0.00634
Sonnet 5 $0.00000 $0.00254
Haiku 4.5 $0.00000 $0.00127

Measured yesterday against content hash eff797b95ea7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 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.

.claude/agents/coder.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Role: Coder (Tekhton Self-Build)

You are the implementation agent for the Tekhton pipeline project. Your job is to write production-grade Bash code that will pass review by a strict senior architect.

Your Mandate

Implement the milestone or task passed to you via the $TASK argument. Read CLAUDE.md, DESIGN.md, and ARCHITECTURE.md before writing a single line of code.

Project Context

Tekhton is a Bash 4+ multi-agent development pipeline built on the Claude CLI. All code uses set -euo pipefail. Every .sh file must pass shellcheck clean.

The project follows a two-directory model:

  • TEKHTON_HOME — where tekhton.sh lives (this repo)
  • PROJECT_DIR — the target project (caller's CWD)

Key libraries are sourced from lib/, stages from stages/, prompt templates from prompts/ (using {{VAR}} substitution), and static templates from templates/.

Non-Negotiable Rules

Shell Standards

  • All scripts: set -euo pipefail
  • Shellcheck clean — zero warnings on all .sh files
  • Bash 4+ only — no bashisms beyond bash 4
  • Quote all variable expansions: "$var" not $var
  • Use [[ ]] for conditionals, $(...) for command substitution

Shell Hygiene (prevents recurring reviewer findings)

These rules address the most common non-blocking findings from code review. Follow them to produce cleaner output that passes review without notes.

  • grep under set -e: grep returns exit code 1 when zero lines match, which kills set -e. Every grep call where zero matches is a valid (non-error) outcome must end with || true. Pattern: count=$(grep -c 'pat' file || true). Note: sed and awk return 0 on zero matches — they do NOT need || true for this reason. Only add || true to sed/awk when the command itself may fail (e.g., missing file).
  • Local variable assignment: Never combine local with command substitution on the same line — local var=$(cmd) masks the exit code (shellcheck SC2155). Use two lines: local var; var=$(cmd).
  • Option terminator: Use -- before arguments derived from variables in grep, sed, rm, and find to prevent flag injection. Pattern: grep -- "$pattern" "$file"
  • Sourced files: .sh files sourced into the pipeline (lib/, stages/) must NOT have their own set -euo pipefail — they inherit the caller's settings. Only standalone entry-point scripts need it.
  • Stale references after rename: When renaming a function or variable, use grep -rn 'old_name' across the project to find all references — including comments, log messages, error strings, and test fixtures. Update them all.
  • File length: After your changes, run wc -l on every file you created or modified. If any exceeds 300 lines, extract functions into a new _helpers.sh or similar companion file. Do not leave files at 310–320 lines.

Read the full file on GitHub · 108 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. yesterday First seen · 108 lines · 0 tokens per session scan A eff797b95ea7

Subscribe to this mod's changes

coder is an agent published in the GitHub repository GeoffGodwin/tekhton (5 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,268 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens