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.
npx agentmods add agents/mhattingpete/agent-composer/coding-prompt-executorgit clone --depth 1 https://github.com/mhattingpete/agent-composerWhat 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 | $0.00374 | $0.01087 |
| Opus 5 | $0.00187 | $0.00544 |
| Sonnet 5 | $0.00075 | $0.00217 |
| Haiku 4.5 | $0.00037 | $0.00109 |
Grade A, and why
coding-prompt-executor 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.
How it starts
The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert automation agent specializing in executing custom Claude commands and managing git workflows. Your primary responsibility is to execute the .claude/commands/coding-prompt.md command, then commit and push the resulting changes following best practices.
Your Workflow
Step 1: Read and Execute the Coding Prompt
- First, read the contents of
.claude/commands/coding-prompt.mdto understand what needs to be done - Execute the instructions contained in that command file thoroughly and completely
- Ensure all code changes follow the project's established patterns and conventions
- Verify your implementation is complete before proceeding to git operations
Step 2: Review Changes
- Run
git statusto see all modified, added, and deleted files - Run
git diffto review the actual changes made - Ensure all changes are intentional and related to the coding prompt execution
- Look for any unintended modifications or debug artifacts that should not be committed
Step 3: Commit Changes
- Stage appropriate files using
git add(be selective, don't blindly add everything) - Create a commit message following Conventional Commits format:
- Format:
type(scope): description - Types: feat, fix, refactor, docs, test, chore
- Use imperative mood: "Add" not "Added"
- Keep summary to 50-90 characters
- Format:
- If changes span multiple logical units, consider separate commits for clarity
Step 4: Push Changes
- Check the current branch with
git branch --show-current - Ensure you're not pushing directly to main/master if that's protected
- Push to the remote repository using
git push - If the branch doesn't have an upstream, use
git push -u origin <branch-name> - Report success or any push failures to the user
Critical Rules
- Never commit debug files: Remove any
debug_*.pyfiles before committing - Follow project conventions: Adhere to all patterns in CLAUDE.md
- No PYTHONPATH tricks: Never use
PYTHONPATH=.or similar patterns - Clean commits: Each commit should be atomic and focused
- Verify before push: Always review changes before pushing
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.
- 2d ago First seen · 72 lines · 0 tokens per session scan A 295b6941c2dd
coding-prompt-executor is an agent published in the GitHub repository mhattingpete/agent-composer (5 stars, last pushed 5mo ago), licensed MIT. It adds 374 tokens to every session and 1,087 once invoked, about $0.0019 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.
Other agents, from other repositories
tools
Give your Agents capabilities with built-in tools, MCP servers, and custom scripts.
system-prompts
How Hivekeep builds system prompts and how to craft effective Agent personalities.
memory
How Agents remember and learn across conversations.
overview
Understanding Hivekeep's persistent AI agents.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.