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.
git clone --depth 1 https://github.com/HigorAlves/orcWrote 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/commands/higoralves/orc/evidence)<a href="https://agentmods.dev/commands/higoralves/orc/evidence"><img src="https://agentmods.dev/badge/commands/higoralves/orc/evidence/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/commands/higoralves/orc/evidence"><img src="https://agentmods.dev/badge/commands/higoralves/orc/evidence.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00047 | $0.01773 |
| Opus 5.5 | $0.00019 | $0.00709 |
| Sonnet 5.5 | $0.00009 | $0.00355 |
| Haiku 4.5 | $0.00005 | $0.00177 |
Grade A, and why
evidence 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 13d 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.
- Bash(curl:*) How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/orc:evidence
Collect browser evidence that a ticket's behavior works — navigate, screenshot, record — then deliver it to the ticket or keep it local. Unlike /orc:qa (a pre-PR gate scoped to your diff that also runs tests/lint), this is scoped to a ticket (or a flow you describe) and does only the evidence loop. Use it to attach proof to a ticket, demo a flow, or QA a ticket you didn't necessarily write the code for.
Collection reuses orc:browser-qa (Driver P/A/B); delivery reuses the orc:evidence-publish skill. This command is only the ticket-first orchestrator — it invents no new browser or upload logic.
Arguments
<TICKET-KEY>— the ticket to scope from and (on upload) deliver to. Omit to use the session's boundjiraTicket, or to run context-only.--context "<what to test>"— describe the flow to exercise, instead of (or on top of) pulling it from the ticket.--target <name>— a named QA target from.orc/targets.json(orc:qa-targets):localprovisions the Docker env; a remote target is probed and used as-is. Omitted ⇒ the Target gate asks.--web <url>— app already running at this URL; skip env provisioning.--driver playwright|agent-browser|chrome— browser driver; default is theorc:browser-qadriver gate.playwright= Playwright Test with a stitched step-tagged video;chrome= Claude-in-Chrome, watch live in your browser;agent-browser= headless CLI validator.--no-env— skip Docker provisioning; use--webor a legacy boot.- Workspace flags (
--repos/--repo/--this-repo) — as in/orc:qa.
Workflow
Phase 0 — Context + ticket
!orc-workspace-detect --banner
Context is injected above (ORC_* vars exported; don't re-run detection). Resolve the target ticket: positional <KEY> → session's bound jiraTicket (orc-state get --field jiraTicket, read-for-data per orc:state-protocol) → AskUserQuestion: Provide a key / Run context-only (save locally, no delivery) / Cancel. A key is optional — context-only runs still collect and save the packet, just with no delivery target.
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.
- 13d ago Changed · +9 lines ed203c19a06e
- 15d ago Changed · +1 lines e4267b4b0301
- 1mo ago First seen · 90 lines · 47 tokens per session scan A dcc1a8716d68
evidence is a command published in the GitHub repository HigorAlves/orc (6 stars, last pushed yesterday), licensed MIT. It adds 47 tokens to every session and 1,773 once invoked, about $0.0002 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 2026-08-31.
Other commands, from other repositories
record
Record a browser walkthrough of a URL using Antigravity (agy). Generates .webm video, screenshots, and a report. Auto-converts to MP4 if ffmpeg is available.
scrape
Scrape structured data from a single URL using Antigravity (agy). Returns markdown table or JSON. Saves to docs/agy/scrapes/.
parallax-status
Summarize the current session and run package/native lifecycle diagnostics.
sync
Rebuild this repo's Claude Code config from its committed manifest.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.