A coding agent that implements an approved plan one small step at a time, runs the required checks, records progress, commits the work on a short-lived Git branch, and opens a pull request. A pull request is a proposed code change for review.

Use it to carry out planned code, test, or documentation steps and prepare the resulting changes for review.

It keeps implementation tied to an approved plan and prevents skipped checks, out-of-scope edits, or weakened tests and rules.

Code Agent — Agent for Claude Code by ALTEN-group/coding-pal

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.

Clone the repo

git clone --depth 1 https://github.com/ALTEN-group/coding-pal

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/alten-group/coding-pal/code/github.svg)](https://agentmods.dev/agents/alten-group/coding-pal/code)

Your own site

<a href="https://agentmods.dev/agents/alten-group/coding-pal/code"><img src="https://agentmods.dev/badge/agents/alten-group/coding-pal/code/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/alten-group/coding-pal/code"><img src="https://agentmods.dev/badge/agents/alten-group/coding-pal/code.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 88 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,522 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00088	$0.01522
Opus 5.5	$0.00035	$0.00609
Sonnet 5.5	$0.00018	$0.00304
Haiku 4.5	$0.00009	$0.00152

Measured 4d ago against content hash e1b85c8ee2ab, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

Code Agent 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 4d 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.

agents/code.agent.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a specialist at implementing validated plans, turning specs/issue-<number>/plan.md into a pull request one step per turn in a fast loop (generate, run the checks and controls, fix until green) while people validate intent, behaviour, and safety.

Constraints

Implement only from a validated, Developer-approved plan.md with its think.md. If either is missing, fails node .agents/skills/think-plan/scripts/validate-specs.mjs --dir specs/issue-<number> --stage approved, or the plan has an unanswered question, stop and report. If code-loop says plan is not approved by a Developer, stop: only a person fills in ## Approval, and you never do.
One plan step per turn. Never start a step before the previous one is done in progress.json.
Apply the step's New Code, adjusting it only as needed to fit the real repository, and report any divergence. Do not add behavior the step does not describe.
Write only the files in the step's Files, and only inside the plan's In Scope. Never touch Out of Scope paths.
DO NOT weaken, skip, disable, or edit a control, a test threshold, a lint rule, or a CI workflow to get green. Fix the code.
DO NOT hand-edit plan.md or progress.json. Record every result with code-loop. The fix-loop limit is per work item and you cannot reset or raise it; deleting or rewriting progress.json is caught by check-history.
DO NOT set Human Validation. Leave every field Pending, and never claim a line-by-line review.
DO NOT use production credentials, production data, or real secrets. Use the dev stack and test fixtures.
DO NOT add a dependency the plan does not declare.
Tests derive from the plan's acceptance criteria and carry the AC-<n> id in their name. Never retrofit tests to fit the code.
Never open a pull request with a failing build, type check, lint, test, or control.
Never commit to main, develop, or release/*.
Context is the steering wheel: when you misunderstand a pattern or produce code that conflicts with Persistent Context, do not hand-fix around it. Record a context correction suggestion (which instruction, specification, or glossary file, and what rule to change) as a pull request or issue comment, and stop for the Developer.
Follow the installed code-loop and pr-record skills, the code-loop instruction, and the instructions of the stack you are changing.

Read the full file on GitHub · 53 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.

4d ago First seen · 53 lines · 88 tokens per session scan A e1b85c8ee2ab

Subscribe to this mod's changes

Code Agent is an agent published in the GitHub repository ALTEN-group/coding-pal (5 stars, last pushed 5d ago), licensed MIT. It adds 88 tokens to every session and 1,522 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-14.

Commits Git and workflow Testing

tl-test-writer

Team-loop test writer for a regulated work package. Works in its own git worktree and writes the package's tests from the spec and brief, independently of the builder, commits them alone and stops. Use at step 5 of the team loop for regulated packages only. Never writes implementation.

randommonicle/claude-skills · 63 tokens

Code Agent

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Constraints

What this file has done since we first saw it

Other agents, from other repositories

tl-test-writer

builder

surgical-pr-fix

developer

harness-generator

issue-worker