Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add archugunov/pm-job-search/plugin install pm-job-searchWrote 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/archugunov/pm-job-search/today)<a href="https://agentmods.dev/skills/archugunov/pm-job-search/today"><img src="https://agentmods.dev/badge/skills/archugunov/pm-job-search/today/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/archugunov/pm-job-search/today"><img src="https://agentmods.dev/badge/skills/archugunov/pm-job-search/today.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00204 | $0.10615 |
| Opus 5 | $0.00102 | $0.05307 |
| Sonnet 5 | $0.00041 | $0.02123 |
| Haiku 4.5 | $0.00020 | $0.01061 |
Grade A, and why
today 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 12d 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 — 487 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/today — daily brief
Produce a short (~half a screen) snapshot that combines: where the user is against their stated strategy, what to actually do next, and a sorted view of the pipeline. Save the brief, regenerate the applications.md index, then print the brief to chat.
Voice: all chat output and drafted content (the daily brief, summary lines, warnings) follows ${CLAUDE_PLUGIN_ROOT}/TONE.md.
Input phase (runs before brief generation)
Capture fresh facts since the last journal entry. Each confirmed fact gets appended to userdata/journal.md and — when it has structured implications — pushed to the relevant userdata/companies/<Co>/meta.md. Then the output phase runs against the just-written state.
This phase always runs (even with no integrations wired) but degrades gracefully — inference and confirms are skipped when there is nothing to scan, and the catch-all prompt becomes the only input surface.
First-run detection (TONE Rule C — don't ask about prior state on first run):
Before prompting for updates, check userdata/journal.md. If the file does not exist, is empty, or contains no entries, skip the entire input phase and go straight to output generation. Do not ask "anything that moved since last time?" — there is no last time.
Step 1 — determine the input window
Read userdata/journal.md. Find the most recent dated heading (## YYYY-MM-DD).
- Gap < 24h → window = 24h. Print nothing; this is the normal daily case.
- Gap 1–6 days → window = the actual gap. Print one line:
Last entry was N days ago — pulling the full window. - Gap ≥ 7 days → window = the actual gap. Print:
You've been away N days — heavier sweep ahead, may take a moment. journal.mdabsent or empty → window = 24h. Treat as a fresh install. Do not print anything special.
Step 2 — inference pass (when integrations wired)
Read userdata/integrations.md to find which MCPs are wired. For each wired integration, scan the window. Same auth-error rule as the existing Integration-data section: surface one line per failed integration at the top of the input phase, do not silently skip, continue with the others.
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.
- 12d ago First seen · 487 lines · 0 tokens per session scan A b828a8748adb
today is a skill published in the GitHub repository archugunov/pm-job-search (7 stars, last pushed 16d ago), licensed MIT. It adds 204 tokens to every session and 10,615 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 2026-08-31.
Other skills, from other repositories
alive:create-walnut
Something new is emerging. A venture, an experiment, a person entering the orbit, a life area getting serious. It needs its own walnut — its own identity, history, and future. Scaffolds the full structure, maps existing context sources, and optionally migrates files across.
alive:bundle
Create, share, and graduate bundles — the unit of focused work within a walnut. Manages the full bundle lifecycle from creation through sharing to graduation.
alive-bundle
Create, manage, and graduate bundles -- the unit of focused work within a walnut.
alive-create
Scaffold a new walnut -- venture, experiment, person, life area, project.
alive-project
Regenerate now.json projections for all walnuts (mechanical, no approval needed).
pilot
One autonomous Pilot cycle - fetch the agenda, execute the top item, journal, exit. Injected by the terminal host - not for manual invocation loops.