pstack is a collection of plain Agent Skills and engineering principles that guide AI-assisted software development across coding agents such as Claude Code, Codex, Pi, Cursor, and OpenCode. It is for developers who want reusable practices for tasks including testing, technical writing, debugging, review, and coordinating agent work. The catalogue entries are the project's skills, agents, and plugin.
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 backnotprop/pstack --skill arenagit clone --depth 1 https://github.com/backnotprop/pstackWrote 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/backnotprop/pstack/arena)<a href="https://agentmods.dev/skills/backnotprop/pstack/arena"><img src="https://agentmods.dev/badge/skills/backnotprop/pstack/arena/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/backnotprop/pstack/arena"><img src="https://agentmods.dev/badge/skills/backnotprop/pstack/arena.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
- Socket pass
- Snyk 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 | $0.00058 | $0.01086 |
| Opus 5.5 | $0.00023 | $0.00434 |
| Sonnet 5.5 | $0.00012 | $0.00217 |
| Haiku 4.5 | $0.00006 | $0.00109 |
Grade A, and why
arena 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 2d 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Arena
Open a todolist with one entry per phase (Frame, Fan out, Cross-judge, Pick, Graft, Verify) before launching anything.
Phase A: Frame
Every candidate gets the same prompt, so the prompt is the contract.
- State the artifact each candidate is producing.
- Derive the rubric, 3-6 concrete gradeable criteria for what success looks like on this task. It is the picker's tool in Phase D. Candidates only see the task.
- Pick the runners. Use the
arena runnersline in the pstack settings file (~/.cursor/rules/pstack-models.mdcin Cursor,~/.claude/rules/pstack-models.mdin Claude Code,~/.agents/pstack-models.mdin other harnesses). If the file or that line is missing, default to one each onclaude-opus-5-5-xhighandgrok-4.7-xhigh-fast. Anautoorinherit-parententry in this line or the cross-judge line means the parent model, so omitmodelfor it. If your subagent tool rejects a configured entry, run that seat on its family's default and say so. Families go by prefix:claude-*, andgrok-*withcursor-grok-*andcomposer-*. A refusedcomposer-*entry runs ongrok-4.7-xhigh-fast. With no family match, useclaude-opus-5-5-xhigh. If it rejects a default, use the closest valid slug of the same family from its error message or your harness's model list. Spawn more runners when the arena covers multiple design directions. Use the same model N times when the work is generation-bound rather than judgment-sensitive. Pick each subagent's reasoning level and fast or standard speed for the job, per the Subagents section of the poteto-mode skill. - Give each candidate its own output location (a git worktree where possible, otherwise
/tmp/arena-<slug>/candidate-<n>/), per the separate-before-serializing-shared-state principle skill.
Phase B: Fan out
Spawn all N subagents in one message with run_in_background: true, each with the task, the path to the shared grounding, its own output path, and instructions to produce the artifact and a short rationale naming the alternatives it considered and what it rejected.
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.
- 2d ago Changed d9ba4883b0fd
- 7d ago Changed 8e68f7f39174
- 10d ago Changed · -23 lines 6c54c51f7115
- 14d ago Changed 031af78a3193
- 18d ago Changed ac9f6f198efd
- 1mo ago Changed 6e8e034aa11c
- 1mo ago First seen · 72 lines · 58 tokens per session scan A a2241e8500a4
arena is a skill published in the GitHub repository backnotprop/pstack (2,580 stars, last pushed 3d ago), licensed MIT. It adds 58 tokens to every session and 1,086 once invoked, about $0.0002 per session on Opus 5.5. 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
email-domain-security
Rigorous, defensible email-spoofability verdict and SPF supply-chain risk analysis computed from published DNS alone. Deepens the record-level SPF/DMARC/DKIM/BIMI/MTA-STS/DNSSEC fetch recipes in the offensive-osint arsenal (§16.14) with the reasoning that section doesn't do: a priority-ordered composite verdict for…
exposure-risk-quantification
FAIR-aligned exposure quantification: turns a pile of recon findings into a defensible 0-100 + A-F org risk score (Likelihood x Impact, three ownership-aware factors: exposure/threat/impact), an ownership + proof demotion cap so unproven or weakly-owned findings can't inflate the number, a $-denominated FAIR…
osint-methodology
Comprehensive OSINT methodology for external red-team operations and authorized attack-surface assessments. Covers the 6-stage recon pipeline (seed → asset expansion → enrichment → exposure analysis → convergence → operator-armed active validation) with connector-resilience and stage-vs-gating discipline, asset-graph…
osint-autopilot
End-to-end external OSINT engagement autopilot. Run the FULL osint-methodology pipeline to completion in ONE go for an authorized domain — engagement folder, Stages 1-5 (seed, expansion, enrichment, exposure, convergence), multi-agent per-host content+JS fan-out, headline verification, auto-generated findings, and a…
analyze-tensegrity-system
Analyze a tensegrity system by identifying compression struts and tension cables, classifying type (class 1/2, biological/architectural), computing prestress equilibrium, verifying stability via Maxwell's rigidity criterion, and mapping biological tensegrity (microtubules, actin, intermediate filaments). Use when…
assess-context
AI context assessment — evaluating problem malleability, mapping structural rigidity versus flexibility, analyzing transformation pressure, and estimating capacity to adapt. Use when a complex task feels stuck and it is unclear whether to push through or pivot, before a significant approach change to assess whether…