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 kdr/overcast --skill overcast-audio-matchgit clone --depth 1 https://github.com/kdr/overcastWrote 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/kdr/overcast/overcast-audio-match)<a href="https://agentmods.dev/skills/kdr/overcast/overcast-audio-match"><img src="https://agentmods.dev/badge/skills/kdr/overcast/overcast-audio-match.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00064 | $0.01049 |
| Opus 5 | $0.00032 | $0.00524 |
| Sonnet 5 | $0.00013 | $0.00210 |
| Haiku 4.5 | $0.00006 | $0.00105 |
Grade A, and why
overcast-audio-match 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 8d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
overcast-audio-match
Use this skill to answer "is this the SAME recording?": Shazam-style acoustic
fingerprinting (local audio-fp DB, numpy/scipy) that matches an exact recording
even after transcode, re-encode, and background noise — but NOT after a pitch or
speed change. Say that twice, because it defines what a match means. Use the broad
overcast skill and overcast/reference/verbs.md for exact flags. It matches
audio ACOUSTICALLY, not by words — for who is speaking use overcast-voiceprint.
Prerequisites
overcast doctor --json # confirm uv + visual-db (numpy/scipy) are ready
scripts/visual-db-uv.sh --audio # install scipy for the fingerprint DB (once per machine)
overcast case init --json
overcast index create audio --type audio-fp --local --json
Workflow
- Fingerprint the known recordings into the index:
overcast audio add ./original-broadcast.mp3 --index audio --json
overcast audio add ./known-song.wav --index audio --json
- Match a query clip against the index, or compare two clips directly. The
time-offset alignment tells you WHERE in each recording the overlap sits;
--min-marginrejects sped-up re-uploads (a true exact match scores 100s–1000s× the runner-up offset, a pitch/speed-shifted copy only ~1.2–1.7×), and--drawrenders an SVG alignment plot (hash-pair scatter + offset histogram) that embeds in briefs likeimage --draw:
overcast audio match ./clip-from-somewhere.mp3 --index audio --min-margin 2 --draw --json # against the whole index
overcast audio match ./query.mp3 ./reference.wav --min-margin 2 --json # clip-to-clip
- Escalate a fingerprint MISS you still suspect is a re-edit. Fingerprinting won't
catch a pitch/speed-shifted or re-performed copy — for that, run a CLAP semantic
pass (
similar, LAION CLAP over abasic-clapindex), which finds acoustically SIMILAR audio rather than the exact recording:
overcast index create audio-sem --type basic-clap --local --json
overcast similar add ./original-broadcast.mp3 --index audio-sem --json
overcast similar match ./clip-from-somewhere.mp3 --index audio-sem --json # semantically nearest audio
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.
- 8d ago First seen · 88 lines · 64 tokens per session scan A aa791c975ba3
overcast-audio-match is a skill published in the GitHub repository kdr/overcast (16 stars, last pushed 5d ago), licensed Apache-2.0. It adds 64 tokens to every session and 1,049 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 2026-08-30.
Other skills, from other repositories
video-to-skill
Turn a video, tutorial, playlist, or course into an installed, evidence-grounded course Skill that can teach, give practice and feedback, apply demonstrated methods, and answer reference questions. Use when the user provides video sources and wants reusable learning or operational capability.
frame-light-leak-cinema
Film light leaks, grain, 16:9 letterbox, and large serif type for cinematic openings or chapter cards.
osmedeus-expert
Expert guide for the Osmedeus security automation workflow engine. Use when: (1) writing or editing YAML workflows (modules and flows), (2) running osmedeus CLI commands (scan, workflow management, installation, server), (3) configuring steps, runners, triggers, or template variables, (4) debugging workflow execution…
recon-tool-integration
Adding a new tool to the recon pipeline: the enrichment-module contract and its isolated wrapper (the actual fan-out and test call path), graph completeness, and the preset catalog that silently strips unknown settings. Miss the isolated wrapper and the tool never runs in parallel; miss the catalog and AI presets drop…
supply-chain-scan
Working on RedAmon's supply-chain scanner (offline OSV + GuardDog + retire + trufflehog): the offline OSV database that the scan path does not bootstrap, the world-readable requirement for the hardened scanner, and the soft-error markers that record what was never analysed. Trigger: editing…
frame-glitch-title
A design template for a glitch-style title frame in Chinese, intended for a video transition, hero image, or cyberpunk graphic. It describes dark backgrounds, distorted lettering, colored offsets, scan lines, noise, and timecode details.