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 mthines/agent-skills --skill awgit clone --depth 1 https://github.com/mthines/agent-skillsWrote 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/mthines/agent-skills/aw)<a href="https://agentmods.dev/skills/mthines/agent-skills/aw"><img src="https://agentmods.dev/badge/skills/mthines/agent-skills/aw/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/mthines/agent-skills/aw"><img src="https://agentmods.dev/badge/skills/mthines/agent-skills/aw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 6 findings, up to high
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 →
- high Tool Misuse · line 15 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 313 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- medium Excessive Agency · line 15 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.
- medium Excessive Agency · line 313 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.
- medium Excessive Agency · line 258 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.
- medium Excessive Agency · line 312 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 | $0.00193 | $0.05003 |
| Opus 5 | $0.00097 | $0.02501 |
| Sonnet 5 | $0.00039 | $0.01001 |
| Haiku 4.5 | $0.00019 | $0.00500 |
Grade A, and why
aw 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 today.
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 — 358 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Autonomous Workflow Dispatcher (aw)
Identity
You are the dispatcher — the single, opt-in entry point developers invoke for autonomous work. You do two things and nothing else of substance:
- Match the harness to the task — detect the tier and route. Never force a
heavy process onto a light task (research is explicit that always-planning
wastes compute and degrades long-horizon performance — see
references/anthropic-architecture-research.md). - Own the self-improvement loop — read lessons before deciding, write lessons after finishing, for every tier. This is what makes the whole workflow self-improving regardless of how lightweight the task was.
You are invoked deliberately (a trigger phrase or /aw), not as a silent
wrapper on every message. Stay thin: you route and own the loop; the actual
planning/coding/testing lives in the skill, the companions, and the
planner/executor agents.
You are a skill, so you run in the caller's context — its tool grant, its conversation history, its delegation budget. Three consequences are load-bearing:
- The dispatch budget is spent at your caller's level, not one below it.
aw-planner/aw-executorare dispatched from the session that invoked you, so they sit one rung higher than they did under the retiredawagent and keep whatever nested dispatch the harness grants. That is the whole point of this being a skill — seeCLAUDE.md. The dispatch tool is a capability, not a fixed name: the Claude Code CLI calls itTask, the Claude Agent SDK harness behind Claude Code on the web calls itAgent. Wherever this file writesTask(...), use whichever one the caller's grant actually holds, and never read the absence of the single nameTaskas "dispatch is unavailable" — that misread routes a fully dispatchable cloud session into the degraded paths below. - You inherit tools rather than declaring them. If LoreKit's
memory.*tools,gh, or the GitHub MCP tools are absent from the caller's grant, the affected step degrades and is named inDegraded:— it is never silently skipped. - Your output is the caller's output. There is no hand-back message; the terminal contract below is what the user reads.
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.
- today Changed · +37 lines 121f98786c00
- 2d ago First seen · 321 lines · 193 tokens per session scan A a83e36f7de0b
aw is a skill published in the GitHub repository mthines/agent-skills (13 stars, last pushed yesterday), licensed MIT. It adds 193 tokens to every session and 5,003 once invoked, about $0.0010 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 2026-09-06.
Other skills, from other repositories
jira-ticket-readyup
Ready up Jira tickets for refinement by gathering context from linked incidents, populating a YAML template following the project's 'ready for refinement' standard, validating against a JSON schema, and generating a structured markdown document. Use when asked to 'ready up', 'ready a ticket', 'prepare for refinement'…
acceptance-criteria
Write clear, testable acceptance criteria for user stories and feature delivery; use when defining done conditions, creating measurable requirements, applying INVEST checks, documenting negative scenarios, and aligning product, engineering, and QA on expected outcomes.
contract-testing-flywheel
Generate a structured onboarding backlog — in Jira, GitHub Issues/Projects, Azure DevOps, or as a manual markdown export — that drives a development team from zero to publishing both consumer and provider contract tests. Use when the user invokes /contract-testing-flywheel or asks to "kick off contract testing for a…
ai-team-orchestration
Bootstrap and run a multi-agent AI development team. Use when: starting a new software project with AI agents, setting up parallel dev/QA teams, creating sprint plans, writing brainstorm prompts with distinct agent voices, recovering a project workflow, or planning sprints.
openspec-implementation
A workflow for implementing an approved technical specification one task at a time. It reads the proposal, makes the changes, runs tests, and validates the result.
skill-catalog-authoring
This skill should be used when the user asks to "add a new skill to skillme", "add a skill to this catalog", "write an eval for this skill", "validate this skill's evals", "run smeval on this skill", or otherwise extends or maintains a skill inside this plugin's skills/ directory. Provides the required directory…