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 yimwoo/hotl-plugin --skill resuminggit clone --depth 1 https://github.com/yimwoo/hotl-pluginWrote 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/yimwoo/hotl-plugin/resuming)<a href="https://agentmods.dev/skills/yimwoo/hotl-plugin/resuming"><img src="https://agentmods.dev/badge/skills/yimwoo/hotl-plugin/resuming.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 5.1 | $0.00027 | $0.02080 |
| Opus 5 | $0.00014 | $0.01040 |
| Sonnet 5 | $0.00005 | $0.00416 |
| Haiku 4.5 | $0.00003 | $0.00208 |
Grade A, and why
resuming 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 7d 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 — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HOTL Resume
Overview
Resume a previously interrupted workflow run. HOTL persists execution state in a sidecar file (.hotl/state/<run-id>.json). On resume, it uses a verify-first strategy to determine whether the interrupted step already succeeded before continuing.
For governed native runs, call hotl-rt reconcile <run-id> (or the selected driver's equivalent) before resuming. Treat its result as a read-only recommendation and retain this skill's verify-first rules; never infer completion from a host session status.
Announce: "Looking for interrupted HOTL runs..."
Sidecar State
Execution state is persisted at .hotl/state/<run-id>.json. This is the authoritative source of truth — workflow checkboxes are a human-visible mirror that may drift after a crash.
Schema
{
"run_id": "<slug>-<YYYYMMDDTHHMMSSZ>-<12-hex-nonce>",
"workflow_path": "/abs/path/to/execution-root/docs/plans/YYYY-MM-DD-<slug>-workflow.md",
"source_workflow_path": "/abs/path/to/original/workflow.md",
"workflow_slug": "<slug>",
"intent": "<from workflow frontmatter>",
"branch": "<branch name>",
"repo_root": "/abs/path/to/repo",
"execution_root": "/abs/path/to/repo-or-worktree",
"worktree_path": null,
"executor_mode": "loop | executing-plans | subagent",
"started_at": "<ISO 8601>",
"last_update": "<ISO 8601>",
"status": "running | paused | blocked | ready_to_finish | completed | abandoned",
"ownership_required": true,
"controller": {"status": "active", "owner_id": "<stable-controller-id>", "lease_expires_epoch": 0},
"current_step": 3,
"total_steps": 8,
"steps": [
{"number": 1, "name": "<step name>", "status": "done | in_progress | pending | failed | blocked", "attempts": 1},
{"number": 2, "name": "<step name>", "status": "done", "attempts": 2}
],
"last_verify_output": "<captured stdout/stderr from last verify>"
}
Run ID Format
<slug>-<YYYYMMDDTHHMMSSZ>-<12-hex-nonce> (e.g., add-auth-20260320T212315Z-a1b2c3d4e5f6). The semantic workflow slug and UTC start time keep the identifier readable; operating-system entropy keeps simultaneous runs unique across independent execution roots. Legacy timestamp-only run IDs remain readable.
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.
- 7d ago First seen · 148 lines · 27 tokens per session scan A 65e8e9d4436f
resuming is a skill published in the GitHub repository yimwoo/hotl-plugin (26 stars, last pushed 2mo ago), licensed MIT. It adds 27 tokens to every session and 2,080 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 2026-08-30.
Other skills, from other repositories
global_rules
Parent policy layer for approval-first coding workflows under both plain and imperial registers. Defines non-negotiable constraints, not the primary auto-trigger workflow.
approved-checklist-executor
Use only after a checklist or atomic slices have been explicitly approved. Execute approved slices under the current run mode, preserve atomic discipline, verify, archive one commit per slice, report git state, package evidence for review, and stop on blockers or replanning needs.
approval-first-planner
Use for implementation-style tasks that modify code, files, pages, or project structure. Inspect first, then output an approval-first execution contract with atomic slices, red lines, green tests, acceptance ladder, commit plan, target artifacts, and a white-box acceptance YAML bridge. Do not edit files.
free-development-mode
A workflow for carrying out coding work after the user explicitly activates free development mode. It allows pragmatic direct execution while keeping scope, safety boundaries, and engineering discipline explicit.
auditor-succession-prompt
A prompt package for creating a fresh auditor role under the Seven-Stars protocol. It defines the auditor’s identity, responsibilities, evidence boundaries, and limits, but does not perform the audit itself.
contract-hook-router
Route host-native plan/ask modes into Cyber-Ming contract lifecycle hooks. Use when a user asks to plan, execute, continue, take over, or when host plan mode should become an Atomic Execution Contract instead of free-form agent planning.