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/dimetron/pi-gonpx agentmods add skills/dimetron/pi-go/vhs-e2e-gifWrote 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/dimetron/pi-go/vhs-e2e-gif)<a href="https://agentmods.dev/skills/dimetron/pi-go/vhs-e2e-gif"><img src="https://agentmods.dev/badge/skills/dimetron/pi-go/vhs-e2e-gif/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/dimetron/pi-go/vhs-e2e-gif"><img src="https://agentmods.dev/badge/skills/dimetron/pi-go/vhs-e2e-gif.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 9 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00106 | $0.01938 |
| Opus 5.5 | $0.00042 | $0.00775 |
| Sonnet 5.5 | $0.00021 | $0.00388 |
| Haiku 4.5 | $0.00011 | $0.00194 |
Grade A, and why
vhs-e2e-gif scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recording an e2e run as a GIF on a PR
Turn a test run into a short GIF and hang it off the PR, so a reviewer sees the suite go green without checking out the branch.
The one trap
VHS segfaults inside the agent sandbox. It binds a loopback port for ttyd,
the sandbox refuses, and randomPort() dereferences the nil listener:
panic: runtime error: invalid memory address or nil pointer dereference
main.randomPort()
github.com/charmbracelet/vhs/tty.go:22
That is the sandbox, not a broken tape. Run every vhs invocation with the
sandbox disabled. Same for gh (network) — the upload step needs it too.
Dependencies: vhs, ttyd, ffmpeg (brew install vhs ttyd ffmpeg). VHS pulls
in the other two, but check before blaming a tape.
Workflow
1. Record.
OUT=/tmp/e2e-tool-calls.gif DIR=$PWD \
scripts/record-gif.sh "go test -tags e2e ./internal/agent/ -run TestE2E -v 2>&1 | grep -E '^(--- |ok|FAIL)'"
The script writes the tape, runs VHS, and extracts the final frame next to
the GIF as <name>.last.png.
2. Read the last frame before posting. A GIF is opaque to you — the PNG is not. Open it and confirm the run actually passed. Posting a recording of a failing suite as evidence of a passing one is the failure mode this step exists to prevent.
3. Attach.
CAPTIONS='e2e: tool calls after the fix' \
scripts/attach-gif-to-pr.sh 246 /tmp/e2e-tool-calls.gif
Uploads as a release asset — never a repo commit — and posts, or edits, a single recording comment on the PR, so a re-record updates that comment in place instead of adding another.
Choosing what to record
Record the narrowest command that proves the claim, filtered to result lines.
A full go test ./... scroll is unreadable at 14pt and the interesting line
scrolls off. These read well:
go test -tags e2e ./internal/agent/ -run TestE2E -v 2>&1 | grep -E '^(--- |ok|FAIL)'
go test -tags e2e ./... 2>&1 | grep -vE '(no test files)$'
make lint
What ships with it
2 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 First seen · 163 lines · 106 tokens per session scan A 5fdc5434ed78
vhs-e2e-gif is a skill published in the GitHub repository dimetron/pi-go (217 stars, last pushed today), licensed MIT. It adds 106 tokens to every session and 1,938 once invoked, about $0.0004 per session on Opus 5.5. 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
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
gh-find-prs
Survey open Codewhale PRs and triage each for mergeability and disposition against the real landing branch.
cw-gates
Use before claiming any Codewhale change is done, green, or ready to land: the focused-to-broad verification ladder, the budget checks CI enforces, and the rules for what counts as a passing test.
triage
Gatekeep and review GitHub issues and pull requests for Qwen Code maintainers. Use for GitHub Action issue triage, PR admission checks, product-direction review, KISS-focused PR review, and staged bilingual GitHub comments.
github
Skill "github" from Hmbown/Codewhale, covering github, when to use, setup, workflow and non-goals.