Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/vibeic/vibe-icnpx agentmods add skills/vibeic/vibe-ic/core-agent-loopWrote 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/vibeic/vibe-ic/core-agent-loop)<a href="https://agentmods.dev/skills/vibeic/vibe-ic/core-agent-loop"><img src="https://agentmods.dev/badge/skills/vibeic/vibe-ic/core-agent-loop/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/vibeic/vibe-ic/core-agent-loop"><img src="https://agentmods.dev/badge/skills/vibeic/vibe-ic/core-agent-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 7 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 173 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 455 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 626 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 627 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 628 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 Data Exfiltration · line 588 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 586 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00224 | $0.11713 |
| Opus 5.5 | $0.00090 | $0.04685 |
| Sonnet 5.5 | $0.00045 | $0.02343 |
| Haiku 4.5 | $0.00022 | $0.01171 |
Grade A, and why
core-agent-loop scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
1. `gh issue view <num>` (or curl + jq) to read body + comments. How it starts
The opening of the file, as written. The whole thing — 806 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Case-study notation. This skill uses
IC-A / USB-HID tester / BENCH-Aas the canonical example chip — substitute your own IC name and host-tester name. The rules themselves are chip-AGNOSTIC and apply to any IC of the matchingic_class(seevibe-ic-marketplace/plugins/vibe-ic/programs/ic_class_profile.py). When you adopt this skill on a different IC, swapIC-A→<your IC name>andUSB-HID tester→<your host-tester name>; the structural gates and rule bodies do not depend on those SKUs.
Core-Agent Loop — Closed-Loop Plugin Issue Fixing
Purpose
The core-agent is the fix-verify-and-close half of the Vibe-IC
quality loop. The field-agent (see vibe-ic:field-agent-loop) runs
the plugin against real benchmark IC projects, finds systematic
gaps, and files them as ORGANIC: GitHub issues. The core-agent
picks those up at every cron wake-up, ships a deterministic
chip-AGNOSTIC fix, self-verifies (reproduce + run the full
plugin test suite the CI way), then closes the issue and adds
the core-closed label. CLOSED is the terminal state. The
field-agent is the audit/reopen safety net: each cron tick it
re-checks closed core-closed issues on the REAL benchmark, marks
the good ones field-verified (stays closed), and gh issue reopens any it finds inadequate (which makes the issue actionable
to the core-agent again). This replaces the old
wait-for-verification limbo — that label is RETIRED.
The loop is chip-AGNOSTIC: fixes describe general plugin
behaviour (regex broadens, schema accepts more synonyms, gate
recognises canonical pattern); no fix references IC-A,
BENCH-A, Vendor, usb_hid_tester, aid, or any vendor IC name as
detection logic.
Contribution-layer framing (so Step 3 is not misread as the public model). The DIRECT PUSH in §Step 3 is the Layer-2 maintainer-internal landing method, used while the plugin is built out — it is NOT what external users do. The Layer-1 public contribution model is unchanged and retained: an external contributor files a backlog (a report, no code) or a PR (a fix, with code). This loop serves both — it polls open backlog items (Step 1) AND auto-lands any externally-filed PR through the gatekeeper flow (§per-tick scope). You are the maintainer, so you ship your OWN fixes by direct push; a non-maintainer contributor never pushes to
main.
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago Changed · +6 lines 1186233c5d44
- 1mo ago First seen · 800 lines · 224 tokens per session scan A ce9de0304f86
core-agent-loop is a skill published in the GitHub repository vibeic/vibe-ic (28 stars, last pushed today), licensed Apache-2.0. It adds 224 tokens to every session and 11,713 once invoked, about $0.0009 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
vivado-cdc-audit
A Vivado workflow for reviewing clock-domain crossing, or CDC, reports in FPGA designs. CDC checks whether signals moving between unrelated clock rates are synchronized and transferred safely.
vivado-timing-closure
A Vivado workflow for finding and fixing setup, hold, and pulse-width timing problems in FPGA designs. Timing closure means meeting the required signal timing without changing the design's intended behavior.
vivado-waveform-debug
A Vivado waveform-debugging workflow for examining VCD simulation traces, which record signal values over time. It can query selected signals and time ranges without opening a full Vivado session.
capturing-learnings
Records a real failure-and-fix pair as a project-local lesson, with the specifics that make it recognisable next time. Use when a bug, leakage mistake, or validation error was found and corrected and should not recur. Use when someone asks to save what was learned from an experiment that went wrong.
analog-netlist-crawl
Crawl and analyze post-layout parasitic netlists without running SPICE. Answers "what's the effective resistance from node A to node B across this massive R mesh?", "inside the VREFN mesh, which device pins are electrically farthest apart?", "which nets have the worst coupling?", "where does settling bottleneck?" — by…
analog-verify
Pre-simulation review and Spectre simulation verification for analog circuits. Reviews circuit netlist and testbench, runs simulation, produces margin report. Use after analog-design completes a netlist.