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 yzhao062/anywhere-agents --skill prungit clone --depth 1 https://github.com/yzhao062/anywhere-agentsWrote 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/yzhao062/anywhere-agents/prun)<a href="https://agentmods.dev/skills/yzhao062/anywhere-agents/prun"><img src="https://agentmods.dev/badge/skills/yzhao062/anywhere-agents/prun/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/skills/yzhao062/anywhere-agents/prun"><img src="https://agentmods.dev/badge/skills/yzhao062/anywhere-agents/prun.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00121 | $0.07063 |
| Opus 5 | $0.00060 | $0.03531 |
| Sonnet 5 | $0.00024 | $0.01413 |
| Haiku 4.5 | $0.00012 | $0.00706 |
Grade A, and why
prun 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 3d 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.
A Codex web-fetch unit can use curl. Report the HTTP status per URL so a cloud-vs-local block shows How it starts
The opening of the file, as written. The whole thing — 451 lines — stays where its author put it; the contents beside it link to each section on GitHub.
prun (parallel run)
Overview
prun fans a task out into independent units that run in parallel on separate-quota or in-session
workers, while the Claude session only coordinates. Workers are Codex (codex exec, a separate abundant account, frontier model) and Sonnet
subagents (inside the Claude session). Codex is the prioritized default: its quota is separate
from the Claude plan and its current model (gpt-6 tier) is strong on hard reasoning and code, so
most units go to Codex. Sonnet is reserved for units that need something the Claude session
uniquely provides (see the Executors rule). The coordinator decomposes the task, dispatches the units, gathers
their results, reviews their diffs, and integrates. It never runs a unit itself.
The orchestrator picks the executor per unit; when in doubt, Codex. A Codex unit runs through the separately authenticated Codex/OpenAI account, so the worker run does not draw on the Claude plan at all. A Sonnet unit and the Claude coordinator both consume the current Claude account's quota; the exact split across models and weekly buckets depends on the plan and on active promotions and shifts over time, so check Settings > Usage before relying on any model-specific split. Codex is the default because its worker run is outside the Claude plan; keep Sonnet units targeted because they draw Claude-side quota.
Relationship to the native Workflow tool
The native Workflow tool fans a task out across Claude subagents under a deterministic script, with structured output, judge panels, and resume. A Workflow run counts against the Anthropic plan's usage and rate limits, and its agents use the session model unless the script routes a stage to a different Claude model.
prun has a different quota shape. A Codex unit is dispatched by a shell call to codex exec,
so the worker run uses the separate Codex/OpenAI account. A Sonnet unit and the coordinating
session both draw the current Claude account's quota, so reserve Sonnet for units that need the
Claude session's own tools. The coordinating session also spends a small Anthropic amount while it
decomposes, dispatches, reads results, and integrates.
What ships with it
13 files 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.
- agents/openai.yaml 691 B
- scripts/dispatch-task.ps1 13 KB runs code
- scripts/dispatch-task.sh 14 KB runs code
- scripts/gather.ps1 3.0 KB runs code
- scripts/gather.sh 4.7 KB runs code
- scripts/monitor.ps1 5.5 KB runs code
- scripts/monitor.sh 7.6 KB runs code
- scripts/prun_state.py 29 KB runs code
- scripts/reap-watch.ps1 4.7 KB runs code
- scripts/report-state.ps1 2.2 KB runs code
- scripts/report-state.sh 1.5 KB runs code
- scripts/snapshot-tail.ps1 2.1 KB runs code
- scripts/snapshot-tail.sh 1.4 KB runs code
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.
- 3d ago Changed ad9f6998463b
- 9d ago First seen · 451 lines · 121 tokens per session scan A dbe59ffbcd1c
prun is a skill published in the GitHub repository yzhao062/anywhere-agents (244 stars, last pushed 2d ago), licensed Apache-2.0. It adds 121 tokens to every session and 7,063 once invoked, about $0.0006 per session on Opus 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 2026-08-30.
Other skills, from other repositories
detecting-dependency-confusion
Detect and prevent public-over-private name resolution in npm, PyPI, and Maven.
detecting-typosquatting-packages-in-npm-pypi
Detects typosquatting attacks in npm and PyPI package registries by analyzing package name similarity using Levenshtein distance and other string metrics, examining publish date heuristics to identify recently created packages mimicking established ones, and flagging download count anomalies where suspicious packages…
detecting-typosquatting-packages
Flag misspelled, brandjacked, and typosquatted package names across npm, PyPI, and crates.io before installation using edit-distance, keyboard-proximity, and known-target corpus matching with typomania, OSSGadget, and pypi-scan.
bumblebee
Run Bumblebee supply-chain inventory and exposure scans on macOS/Linux to detect compromised packages, extensions, and MCP host configs.
ship-safe-red-team
Run a multi-agent red team scan — 29 specialized security agents scan for 80+ attack classes including injection, auth bypass, SSRF, supply chain, Supabase RLS, MCP security, agentic AI, RAG poisoning, PII compliance, and more. Use when the user wants a deep security analysis beyond just secrets.
ship-safe-baseline
Manage your security baseline — accept current findings as known debt, then only report new regressions on future scans. Use when the user wants to adopt security scanning incrementally or suppress existing findings.