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 wrg32786/aigent-os --skill resumegit clone --depth 1 https://github.com/wrg32786/aigent-osWrote 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/wrg32786/aigent-os/resume)<a href="https://agentmods.dev/skills/wrg32786/aigent-os/resume"><img src="https://agentmods.dev/badge/skills/wrg32786/aigent-os/resume/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/wrg32786/aigent-os/resume"><img src="https://agentmods.dev/badge/skills/wrg32786/aigent-os/resume.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 Agent Snooping · line 9 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00056 | $0.01099 |
| Opus 5 | $0.00028 | $0.00549 |
| Sonnet 5 | $0.00011 | $0.00220 |
| Haiku 4.5 | $0.00006 | $0.00110 |
Grade A, and why
resume 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 3d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/resume
The entire job: load → re-ground → ACT on waiting_on. Resumption is proven by the action taken, never by capsule text being in context. The verb ends when the first real step from waiting_on/next_valid_action has been taken.
When this skill fires vs the automatic verb
The post-clear boot does NOT need this skill — daemons/resume-verb.mjs injects the full procedure automatically on SessionStart(clear). Invoke /resume for the explicit cases: re-grounding on request without a clear, or picking up work mid-session.
The consume contract (the runtime enforces this; know it, don't re-implement it)
On the post-clear boot, the capsule the runtime loads is marked status: resumed mechanically at load — the same capsule is never silently re-resumed on a later clear. If no active capsule exists, the boot takes the documented degraded path (re-derive from live memory) and says so; it never replays stale state as fresh. An explicit mid-session /resume is a re-ground, not a consume — it does not mark, so say plainly which capsule you re-grounded from.
The rejection ledger (part of the load, not decoration)
Everything the selector skipped is printed in the injected procedure under CAPSULES NOT SELECTED, grouped by reason with counts and example filenames. It exists because a capsule silently discarded and a capsule that never existed look identical from where you are sitting, and only one of those is a defect.
already-consumedis ordinary history: a previous cycle spent that capsule on purpose.- Any other reason means a capsule somebody wrote was thrown away. If a capsule you expected to resume from appears there, the SELECTOR is the bug. Report it; never hand-edit the capsule to satisfy the matcher, because that hides the defect and leaves it in place for the next session.
- A full ledger with nothing selected means every candidate on disk was rejected. That is a defect until proven otherwise, not an empty install.
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.
- 3d ago Changed · +1 lines ff6b704d3d25
- 7d ago First seen · 46 lines · 56 tokens per session scan A dc25492c16dd
resume is a skill published in the GitHub repository wrg32786/aigent-os (18 stars, last pushed yesterday), licensed MIT. It adds 56 tokens to every session and 1,099 once invoked, about $0.0003 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 skills, from other repositories
send-email
Compose and send a one-off email to a named recipient via Resend - written in the operator's voice, then sent in-run through the shared send caps with an operator audit copy.
RemoteTeamCommander
Complete operating system for remote and distributed teams — async communication, time zone equity, virtual culture, performance management at a distance, and preventing remote-class citizens.
chief-of-staff
Act as the user's chief of staff, not a task-taker. Use for planning a day or week, triaging priorities, prepping for meetings or calls, drafting messages on the user's behalf, tracking commitments and follow-ups, reviewing what's in-flight, or any moment the user wants a trusted operator who knows their context.…
meeting-analyzer
Activates MeetingAnalyzer for extracting intelligence from meeting transcripts, notes, and recordings. Use when you need to extract action items with owners and due dates, identify decisions made, summarize key discussion points, detect unresolved conflicts or blockers, or generate a structured meeting follow-up email.
Loop
Run the same task on a recurring cron schedule. Parses interval syntax (5m, 2h, 1d, "/15 "), records the schedule, and acknowledges back to the user.
Batch Run
Apply the same prompt or operation to a list of items in turn, collecting results. Useful for "do X to each of these files / PRs / records" tasks.