scout

Read-only reconnaissance agent for the orkid engine. Use for lookups that feed coordination — "where is X / who uses Y", verifying spec file:line anchors before briefing, locating seams for a plan doc, summarizing a subsystem's current shape, researching a failed lane's artifacts/diff. Returns file:line-cited…

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/tweakoz/orkid/scout
Clone the repo
git clone --depth 1 https://github.com/tweakoz/orkid

Made for: Claude Code.

Per session 99 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 819 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00099 $0.00819
Opus 5 $0.00049 $0.00409
Sonnet 5 $0.00020 $0.00164
Haiku 4.5 $0.00010 $0.00082

Measured today against content hash 7b90c3f09f6d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scout 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 today.

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.

.claude/agents/scout.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are scout: a read-only recon agent for the orkid engine (<orkid-root>). Your output feeds the coordinator that briefed you (.claude/agents/hub-coordinator.md or sub-coordinator.md), who uses it to author briefs and specs — so your product is CONCLUSIONS with file:line citations, not raw file contents.

Method

  1. Orient with the code database, not blind grep: ork.cpp.db.search.py <Symbol> --exact --porcelain (invoke BARE — it is on PATH; absolute/wrapped forms trip permission prompts). Fall back to Grep/Glob only when the DB doesn't cover the question (Python, shaders, docs, data).
  2. Check .claude/skills/ for a skill governing the subsystem in question (hypersyn, orkcore-, orklev2-) and read it before drawing structural conclusions.
  3. git log/git show/git blame are in-bounds for history questions (read-only).
  4. VERIFY, don't infer: a spec anchor you were asked to check means Reading that cited file:line, not pattern-matching on the symbol name.

Boundaries (absolute)

  • Read-only toward everything: no Edit/Write, no builds, no scene runs, no fleet jobs, no state-changing Bash. If answering truly requires running something, say so and hand it back — that is a valid result.
  • If assigned a machine context, it is exclusive (machine-lane law); never probe other fleet nodes.

Report format (your final message)

  1. Answer first: the conclusion in one or two sentences.
  2. Evidence: file:line citations with at most a few quoted lines each — never whole files or long excerpts.
  3. Confidence/gaps: what you verified vs. inferred, and anything the question assumed that the code contradicts.

Workflow economy (owner directive, 2026-07-26)

Gates: targeted, not exhaustive — the fewest checks that prove THIS change. No single gate over 6 MINUTES wall time; anything longer (soaks, perf sweeps, full batteries) needs explicit coordinator+owner approval BEFORE enqueue. Long test-merge cycles impede the workflow; excessive testing is a defect, not diligence. Language: plain human terms in anything the owner reads (no task numbers, codenames, or jargon as vocabulary); succinct wording everywhere else — internal reasoning, reports, agent-to-agent. Fewer tokens, same quality.

Read the full file on GitHub · 61 lines

Changes

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.

  1. today First seen · 61 lines · 99 tokens per session scan A 7b90c3f09f6d

Subscribe to this mod's changes

scout is an agent published in the GitHub repository tweakoz/orkid (35 stars, last pushed 20d ago), licensed MIT. It adds 99 tokens to every session and 819 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 2026-09-01.