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 skills add vladkesler/initrunner --skill reply-composergit clone --depth 1 https://github.com/vladkesler/initrunnerWrote 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/skills/vladkesler/initrunner/reply-composer)<a href="https://agentmods.dev/skills/vladkesler/initrunner/reply-composer"><img src="https://agentmods.dev/badge/skills/vladkesler/initrunner/reply-composer/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/skills/vladkesler/initrunner/reply-composer"><img src="https://agentmods.dev/badge/skills/vladkesler/initrunner/reply-composer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 118 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00032 | $0.00863 |
| Opus 5.5 | $0.00013 | $0.00345 |
| Sonnet 5.5 | $0.00006 | $0.00173 |
| Haiku 4.5 | $0.00003 | $0.00086 |
Grade A, and why
reply-composer 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 1mo 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Email reply composition skill.
When to activate
Use this skill when the user asks to reply to an email, draft a response, or compose a new message.
Process
1. Gather context
Collect everything needed before writing:
- Read the original email with
read_email - If it's a thread, search for earlier messages in the thread:
search_inbox(query='SUBJECT "<thread subject>"') - Recall the sender profile from semantic memory
- Recall any style rules from procedural memory:
recall("style_rule") - Recall any previous replies to this sender:
recall("<sender> reply_sent")
2. Analyze tone
Determine the appropriate tone from the original message:
| Original tone | Reply tone | Indicators |
|---|---|---|
| Formal | Formal | Full sentences, titles, "Dear", "Sincerely" |
| Semi-formal | Semi-formal | Professional but relaxed, first names, "Thanks" |
| Casual | Casual | Short sentences, contractions, emoji, "Hey" |
| Terse | Match length | One-liners, bullet points, no greeting |
Adapt to the sender's style. If the sender writes two-sentence emails, don't reply with five paragraphs.
3. Research if needed
If the reply requires information you don't have:
- Use
searchto look up facts, documentation, or context - Use
web_readerto read referenced URLs from the original email - Recall topic context from semantic memory
4. Compose draft
Write the reply following these rules:
Structure:
- Greeting (match sender's style)
- Acknowledge their message (one sentence, reference specifics)
- Body (answer questions, provide information, address requests)
- Next steps or call to action (if applicable)
- Sign-off (check procedural memory for preferred sign-off)
Thread-aware replies:
- Reference specific points from the thread, not just the last message
- Don't repeat information already established in the thread
- If the thread has multiple participants, address the relevant ones
Length:
- Match the expected length to the complexity of the response
- Simple confirmations: 1-2 sentences
- Detailed answers: structured with bullet points or numbered lists
- Keep under 300 words unless the topic requires more
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.
- 1mo ago First seen · 120 lines · 32 tokens per session scan A 5ba4c28327ee
reply-composer is a skill published in the GitHub repository vladkesler/initrunner (40 stars, last pushed 9d ago), licensed Apache-2.0. It adds 32 tokens to every session and 863 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
Vizra ADK Evaluation Framework
Test and evaluate AI agents with automated evaluations, assertions, and LLM-as-a-Judge patterns.
Vizra ADK Tool Creation
Build custom tools for Vizra ADK agents - includes patterns for database, API, file, and email tools.
Vizra ADK Memory System
Implement persistent memory, session context, and vector memory (RAG) for AI agents.
Vizra ADK Workflows
Orchestrate complex multi-agent workflows - sequential, parallel, conditional, and loop patterns.
Vizra ADK Agent Creation
Create AI agents with Vizra ADK - includes patterns for customer service, data analysis, and content generation agents.
community-skill-radar
Searches Reddit communities for OpenClaw pain points and feature requests, scores them by signal strength, and writes a prioritized PROPOSALS.md for you to review and act on.