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 ivklgn/ai-kit --skill recapgit clone --depth 1 https://github.com/ivklgn/ai-kitWrote 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/ivklgn/ai-kit/recap)<a href="https://agentmods.dev/skills/ivklgn/ai-kit/recap"><img src="https://agentmods.dev/badge/skills/ivklgn/ai-kit/recap/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/ivklgn/ai-kit/recap"><img src="https://agentmods.dev/badge/skills/ivklgn/ai-kit/recap.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 Excessive Agency · line 4 Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
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 | $0.00120 | $0.00967 |
| Opus 5 | $0.00060 | $0.00483 |
| Sonnet 5 | $0.00024 | $0.00193 |
| Haiku 4.5 | $0.00012 | $0.00097 |
Grade A, and why
recap 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recap
Everything needed is already in context. This skill is fast because it refuses to go looking for anything else.
Speed contract
Run exactly one command, and only if the working directory is a git repository:
git status --short --branch && git --no-pager log --oneline -10
If it fails (not a repository), continue silently without it.
Never call Read, Grep, Glob, Edit, WebSearch, or any other tool. Never re-open files that were already discussed — the discussion is in context. Never launch a subagent. One command, then write the answer.
Language
Write the recap in the language the user has been writing in this session, not the language of this skill. Mixed input follows the dominant language of the user's own messages. Translate the slot labels too — in Russian they read Цель / Сделано / Сейчас / Открыто / Дальше. Keep identifiers verbatim — file paths, branch names, commands, flags, library names, error strings.
Level 1 — default output
No argument means the full status. Fill these five slots, in order:
Goal — what the user is actually trying to achieve, in their terms. One line. If the session pivoted, state the current goal and mark the pivot in a clause.
Done — 2–5 bullets, one line each. Each bullet is a result, not an activity: "found the built-in /recap, it is one line long" rather than "searched for skills". Where a decision was made, name the decision and its reason. Merge related steps into one bullet.
Now — the single thing in flight and how far it got. If the last turn completed cleanly, name that completed thing instead.
Open — 0–3 bullets of what is expensive to rediscover: approaches tried and rejected with the reason, unanswered questions, known traps, work deliberately deferred. Omit the whole slot when genuinely empty — never pad it.
Next — the immediate next action, concrete enough to start on. If a user decision is required first, name the decision instead.
Close with a one-line hint that any slot or topic can be expanded.
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 · 65 lines · 120 tokens per session scan A 038234bd0e1f
recap is a skill published in the GitHub repository ivklgn/ai-kit (12 stars, last pushed 7d ago), licensed MIT. It adds 120 tokens to every session and 967 once invoked, about $0.0006 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 2026-09-04.
Other skills, from other repositories
goal-delivery-completeness
Check that delivered work matches the user's goal. Use before final response, release, or handoff.
session-output-contract
Make session output useful, complete, and resumable. Use at handoff, final response, or long-running task summary.
stale-baseline-guard
Avoid acting on outdated context. Use after interruptions, long sessions, branch changes, or concurrent edits.
notion_synthesize
Synthesize durable ktx wiki pages and semantic-layer sources from staged Notion pages, databases, data-source rows, and clustered Notion evidence. Load when a WorkUnit contains Notion raw files or Notion evidence chunks.
codex-skin-pack-installer
Install Codex themes and Codex skin packs with npx skills. Use this skill when a user wants Codex to find, download, validate, apply, switch, remix, or restore Codex desktop themes such as caishen-readable, caishen-lite, mythic-guardian-noir, global-founder-bright, export-night, or caishen-max. It stages verified…
find-journalists
Build, refine, dedupe, and enrich small fit-checked journalist lists for newsjack campaigns. Uses the newsjack CLI (preferred) or the medialyst MCP for news search and journalist enrichment, and falls back to a best-effort local mode with no verified contacts; the agent owns how returned data is organized.