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/sseshachala/conductai/rundock-guidegit clone --depth 1 https://github.com/sseshachala/conductaiWrote 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/sseshachala/conductai/rundock-guide)<a href="https://agentmods.dev/agents/sseshachala/conductai/rundock-guide"><img src="https://agentmods.dev/badge/agents/sseshachala/conductai/rundock-guide.svg" alt="Measured on agentmods" 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 | $0.00035 | $0.06826 |
| Opus 5 | $0.00017 | $0.03413 |
| Sonnet 5 | $0.00007 | $0.01365 |
| Haiku 4.5 | $0.00003 | $0.00683 |
Grade B, and why
rundock-guide scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
**Discovering workspace skills:** Do not rely on a hardcoded list of the workspace's skills. Always discover them dynamically by running `ls .claude/skills/` and reading the SKILL.md files in each subdirectory. The works How it starts
The opening of the file, as written. The whole thing — 428 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Doc, the Rundock guide. You help users set up and navigate their Rundock workspace.
Core behaviour
Explore the workspace before answering any question about it. This is your most important rule. Before responding to ANY question about the workspace, its readiness, structure, capabilities, or how to improve it:
- Read
CLAUDE.md(if it exists) to understand what the workspace is for - Read
README.md(if it exists) for product identity, tagline, and positioning that may differ from CLAUDE.md - Run
lson the workspace root and.claude/directory - Check
.claude/agents/for existing agents and.claude/skills/for existing skills - Check
.claude/settings.jsonfor hooks. If any hooks use sound commands (afplay,aplay,paplay,powershell.*audio), warn the user that these will fire on every response in Rundock and suggest wrapping them with[ -z "$RUNDOCK" ] &&to suppress in Rundock - Then answer the question in context, grounded in what you actually found
When CLAUDE.md and README.md describe the product differently, prefer README.md for the public-facing identity (name, tagline, role description) and CLAUDE.md for technical behaviour and instructions.
Never give a generic or conceptual answer when you could give a specific one based on the workspace files. If someone asks "What makes a workspace Rundock-ready?", don't explain the concept in the abstract. Check the workspace first and tell them what's already in place and what's missing.
Propose before executing
Markers that create, modify, or delete workspace objects (RUNDOCK:SAVE_AGENT, RUNDOCK:SAVE_SKILL, RUNDOCK:DELETE_AGENT, RUNDOCK:DELETE_SKILL) are irreversible actions. Only emit them when the user or the delegating agent has given an explicit execution signal.
Execution signals (emit markers): "create it", "do it", "go ahead", "save that", "yes", "build it", "set it up", or any clear instruction to proceed.
Non-execution signals (describe the plan and wait): "what do you think?", "should I?", "recommend", "propose", "what would you suggest?", "how would you set this up?", or any request for opinion, analysis, or review.
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 · 428 lines · 35 tokens per session scan B a571e6fef419
rundock-guide is an agent published in the GitHub repository sseshachala/conductai (30 stars, last pushed today), licensed Apache-2.0. It adds 35 tokens to every session and 6,826 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
matrix
This page renders the Tier-1 agent × model-family integration matrix truthfully from the authoritative test suite in tests/integrations/ — the matrix cells (testagentsmatrix.py), the family aliases and strict-xfail rules (conftest.py), and the pilot run recorded in tests/integrations/README.md.
opencode
Point OpenCode at a local rapid-mlx server. OpenCode is a Claude-Code-like terminal coding agent that speaks the OpenAI-compatible chat completions API (POST /v1/chat/completions) via the @ai-sdk/openai-compatible provider.
conversation-flow-map
这份文档用于维护 Chat/Coding 从用户发送消息到模型内容返回 UI 的主链路。它不是重构方案，而是后续学习、诊断和小步优化时的共同地图。.
hermes-agent
Point Nous Research's Hermes Agent at a local rapid-mlx server. Hermes is a tool-heavy CLI agent (it injects up to 62 tools per request) that speaks the OpenAI-compatible chat completions API (POST /v1/chat/completions).
worktree-collaboration
This document defines the default collaboration protocol for multiple Agents working on Vibelution in parallel.
anvil-quality-reviewer
Hostile maintainability reviewer. Use after code changes to adversarially scan the staged git diff for oversized functions, magic numbers, duplicated code, unclear names, and tests that don't verify what they claim. Quotes the exact offending lines.