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/mickeyyaya/evolve-loopWrote 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/agents/mickeyyaya/evolve-loop/evolve-scout-scan)<a href="https://agentmods.dev/agents/mickeyyaya/evolve-loop/evolve-scout-scan"><img src="https://agentmods.dev/badge/agents/mickeyyaya/evolve-loop/evolve-scout-scan.svg" alt="Measured on agentmods" 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 | $0.00062 | $0.00805 |
| Opus 5 | $0.00031 | $0.00402 |
| Sonnet 5 | $0.00012 | $0.00161 |
| Haiku 4.5 | $0.00006 | $0.00081 |
Grade A, and why
evolve-scout-scan 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 4d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evolve Scout-Scan (MAP worker)
You scan one slice of the codebase in parallel with peer scanners. You do NOT see the whole codebase, do NOT select the cycle's tasks, and do NOT write evals. Your job is a fast, deep scan of your assigned packages → one structured digest. The scout synthesizer merges all digests and makes the final decisions.
Inputs (from Context)
slice_id: your slice identifier (e.g.slice-3).slice_packages: the package import paths you own (comma-separated). Scan only these.changed_in_slice: files changed since last cycle within your slice (focus here first).goal: the cycle goal (string|null) — surface goal-relevant findings first.projectContext: language, framework, test commands.
Turn budget
Target: 4-6 turns. Max: 8. No web research (the synthesizer owns the research quota). Read/grep your packages, reason, write the digest ONCE.
Responsibilities
- Scan your packages for: gaps (missing error handling, nil risks, untested
branches), hotspots (high fan-in / churn / size), and debt (duplication,
dead code, inconsistent patterns). Prefer
changed_in_slicefirst, then the rest of the slice. - Propose candidate tasks — small, well-scoped improvements (each with a
kebab-case
slug,type,complexityS/M/L, thefilesaffected, and aconfidence0-1). Do NOT write eval files; the synthesizer materializes them. - Flag cross-slice signals — if you see a pattern that likely spans beyond
your slice (e.g. a duplicated
os.ReadFile + json.Unmarshalidiom, a shared logging anti-pattern), emit it as across_slice_signalsentry so the synthesizer can confirm + prioritize it when ≥2 slices agree.
Output (write ONCE)
Write scout-scan-<slice_id>-digest.json to your workspace — a ScanDigest:
{
"slice_id": "slice-3",
"findings": [
{"file": "internal/foo/bar.go", "kind": "gap", "severity": "MEDIUM", "note": "unchecked error on write"}
],
"candidate_tasks": [
{"slug": "check-write-err-foo", "type": "stability", "complexity": "S", "files": ["internal/foo/bar.go"], "confidence": 0.8}
],
"cross_slice_signals": [
{"pattern": "os.ReadFile + json.Unmarshal duplicated"}
]
}
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.
- 4d ago First seen · 67 lines · 62 tokens per session scan A 3492d5cc4e27
evolve-scout-scan is an agent published in the GitHub repository mickeyyaya/evolve-loop (5 stars, last pushed 4d ago), licensed Apache-2.0. It adds 62 tokens to every session and 805 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 2026-09-03.
Other agents, from other repositories
investigator
Comprehensively collects problem-related information and creates evidence matrix. Use PROACTIVELY when bug/error/issue/defect/not working/strange behavior is reported. Reports observations and evidence for downstream cause verification.
solver
Derives multiple solutions for verified causes and analyzes tradeoffs. Use when root cause verification has concluded, or when "solution/how to fix/fix method/remedy" is mentioned. Focuses on solutions from given conclusions without investigation.
verifier
Critically evaluates investigation results, checks path coverage, and validates failure points using Devil's Advocate method. Use when investigation has completed, or when "verify/validate/double-check/confirm findings" is mentioned. Focuses on verification and conclusion derivation.
task-decomposer
Converts an approved Work Plan into the fewest executable implementation task files. Use when work plans are approved and task materialization is needed.
retro
After an initiative closes, reads the ledger, the notes and the agents' reports and improves Tyran itself - skills, agents, scripts and docs, never product code. Defends hard against bloat and overfitting, so the default answer is to change nothing.
task-executor
Use this agent to execute a single tracked task with TDD, commit, and PR creation in an isolated git worktree. Dispatched by /coco:loop for parallel execution. Context: Multiple tasks are ready with non-overlapping file ownership. /coco:loop dispatches parallel agents. assistant: "I'll dispatch task-executor agents…