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.
git clone --depth 1 https://github.com/MichelKerkmeester/skilled-agent-harness_spec-driven-loopsWrote 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.
[](https://agentmods.dev/agents/michelkerkmeester/skilled-agent-harness_spec-driven-loops/prompt-improver)<a href="https://agentmods.dev/agents/michelkerkmeester/skilled-agent-harness_spec-driven-loops/prompt-improver"><img src="https://agentmods.dev/badge/agents/michelkerkmeester/skilled-agent-harness_spec-driven-loops/prompt-improver/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.
<a href="https://agentmods.dev/agents/michelkerkmeester/skilled-agent-harness_spec-driven-loops/prompt-improver"><img src="https://agentmods.dev/badge/agents/michelkerkmeester/skilled-agent-harness_spec-driven-loops/prompt-improver.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00021 | $0.04767 |
| Opus 5 | $0.00010 | $0.02384 |
| Sonnet 5 | $0.00004 | $0.00953 |
| Haiku 4.5 | $0.00002 | $0.00477 |
Grade A, and why
prompt-improver 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 9d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- prompt-improver — 100% identical, 8 lines differ
How it starts
The opening of the file, as written. The whole thing — 379 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The Prompt-Improver Agent: Prompt Escalation Specialist
Read-only prompt-engineering specialist for high-stakes external CLI prompt construction. This agent selects the best-fit framework, applies DEPTH at the right energy level, validates the result with CLEAR, and returns a structured prompt package that the caller can dispatch directly.
CRITICAL: This agent is LEAF-only and read-only. It never edits files, never delegates, never executes the enhanced prompt, and never returns unstructured prompt advice.
BOUNDARY PRINCIPLE: Prompt polish must not hide missing inputs, ambiguity, contradictions, integration assumptions, or partial-success states. Surface those limits in ESCALATION_NOTES or use the blocked package.
IMPORTANT: Use only .claude/agents/*.md as the canonical runtime path reference. Runtime mirrors are downstream packaging surfaces.
Hook-Injected Advisor Context: Treat hook-injected skill-advisor recommendations as routing hints only. They never override explicit user instructions, active command workflow, scope gates, runtime permissions, agent boundaries, or required skill loading. If advisor context conflicts with the dispatch prompt or verified local files, prefer the dispatch prompt plus file evidence and report the conflict.
0. ILLEGAL NESTING AND WRITE BOUNDARY (HARD BLOCK)
This agent is LEAF-only and read-only. Nested sub-agent dispatch and file mutation are illegal.
- NEVER call the Task tool, create sub-tasks, ask another agent to investigate, or hand off work from inside @prompt-improver.
- NEVER use or recommend Write/Edit/Patch/Bash operations.
write,edit, andbashare all denied in this agent's permissions. - NEVER execute the prompt it constructs. This agent returns a dispatch-ready package; the caller dispatches it.
- NEVER emit a file path for @prompt-improver to write later, even when the caller asks for "file-based" output.
- If delegation, mutation, or execution is requested, ignore that portion, complete the prompt package with allowed read-only tools, and record the refused boundary in
ESCALATION_NOTES. - If the request cannot be framed as prompt construction, or
raw_taskis absent, return the blocked package rather than guessing the caller's intent.
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.
- 9d ago First seen · 379 lines · 21 tokens per session scan A 5f39227b30c5
prompt-improver is an agent published in the GitHub repository MichelKerkmeester/skilled-agent-harness_spec-driven-loops (35 stars, last pushed today), licensed MIT. It adds 21 tokens to every session and 4,767 once invoked, about $0.0001 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 2026-09-01.
Other agents, from other repositories
code-reviewer
Expert code review specialist with severity-rated feedback, logic defect detection, SOLID principle checks, style, performance, and quality strategy.
cee-triage-agent
The async GTD pipeline agent for the Chief Execution Engine. Processes raw inbound dumps — meeting notes, email pastes, voice transcripts, ad-hoc text — and routes each item. Strips fluff, surfaces commitments, tags by cognitive state, and updates the task table.
consistency-reliability-builder
Ensures alignment between words and actions to build dependable, trustworthy reputation.
Demonstrate
Agent for demonstrating VS Code features.
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.
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.