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 grayhatdevelopers/vidxp --skill vidxp-find-video-evidencegit clone --depth 1 https://github.com/grayhatdevelopers/vidxpWrote 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/grayhatdevelopers/vidxp/vidxp-find-video-evidence)<a href="https://agentmods.dev/skills/grayhatdevelopers/vidxp/vidxp-find-video-evidence"><img src="https://agentmods.dev/badge/skills/grayhatdevelopers/vidxp/vidxp-find-video-evidence.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 | $0.00098 | $0.00790 |
| Opus 5 | $0.00049 | $0.00395 |
| Sonnet 5 | $0.00020 | $0.00158 |
| Haiku 4.5 | $0.00010 | $0.00079 |
Grade A, and why
vidxp-find-video-evidence 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Find video evidence with VidXP
Workflow
- Resolve the
vidxpMCP tools, then callget_workspace. If the requested video is not indexed, explain that it must be indexed first. - Submit one retrieval job. Use
search_momentsto locate moments; usequery_videoonly when the user asks for a synthesized answer. Usecommand.querywithsearch_momentsandcommand.questionwithquery_video. Setcommand.media_idwhen the user means one video. - In that initial job, put exactly this inside
command:"evidence_delivery": {"mode": "keyframes_and_clips", "max_items": 3}. This prepares the ranked board, standalone keyframes, and clips without a second retrieval pass. Never sendcommand.materialize. - Call
wait_jobfor bounded waits. Pass itsobservation_tokenasafter_observation_tokenon the next wait. When terminal, callget_job_evidenceonce. It returns the concise evidence index and visual content without the full structured job dump. Search and query may take time; update the user when the stage changes or about once per minute, never after every wait and never with an invented ETA. - Surface the returned board, keyframes, and clips immediately. Do not call
get_job, repeat the search, materialize more evidence, create another board, or perform a self-directed verification loop before showing the initial evidence. - Stop after the first evidence delivery. Only when the user explicitly asks
for another selection or format, use tile evidence IDs with
materialize_job_evidence, or usecreate_evidence_boardfor a custom selection ornext_start_rankcontinuation.
Actor scope
- Actor data is available through
query_video, not name search. It represents anonymous, video-scoped face clusters—not a named or cross-video identity. - Treat its image as a representative full frame, not an exact face crop or proof of continuous presence. Do not claim exhaustive named appearances.
- For named-person requests, surface the best scene candidates immediately and label uncertain matches as candidates. Do not delay delivery while trying to prove identity through additional searches.
What ships with it
1 file 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.
- 8d ago First seen · 60 lines · 98 tokens per session scan A 97217fbf97a0
vidxp-find-video-evidence is a skill published in the GitHub repository grayhatdevelopers/vidxp (49 stars, last pushed today), licensed MIT. It adds 98 tokens to every session and 790 once invoked, about $0.0005 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
native-subtitle-quote-image
A tool for turning selected video moments and their subtitles into tall social-media images. It supports either keeping subtitles already embedded in the video or drawing approved text onto real video frames.
reflectworld-percept
Perception automatically analyzes video segments from Eye using cloud VLM (Gemini). Working Memory tracks cross-segment event context. Results flow to ReflectWorld (storage) and Act (rules) via PostProcessor — all automatic.
reflectworld
You have access to ReflectWorld, a visual perception and persistent memory system. It gives you eyes and memory for cameras.
authoring-video-templates
Use when creating or editing a video template JSON (the template descriptor), adding or changing sections/filters/maps/variables/transitions/looks/motion/audio/layers, or debugging template validation errors in ffmpeg-video-composer.
core-architecture-patterns
Use when adding a segment type, platform adapter, editor manager, core service, or descriptor effect (look/grade/motion/section-audio field) in packages/ffmpeg-video-composer, or when wiring new dependencies into the tsyringe container.
reflectworld-mem
Search and manage ReflectWorld video memories from ReflectWorld cameras.