peer-agents

A tool for assigning bounded tasks or independent reviews to Claude Code or the Codex command-line agent. It can run these peer agents in the foreground or as background jobs, with optional access to write in a worktree.

In plain words
What is it for?
Use it for repository reviews, implementation tasks in a separate worktree, or quick technical questions.
Why use it?
It lets another coding agent inspect or work on a focused task using its own subscription budget.

Skill for Claude CodeCodex

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 skills/jonnyton/tinyassets/peer-agents
Any agent
npx skills add Jonnyton/TinyAssets --skill peer-agents
Clone the repo
git clone --depth 1 https://github.com/Jonnyton/TinyAssets

Made for: Claude Code, Codex.

Per session 53 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,659 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00053 $0.01659
Opus 5 $0.00026 $0.00830
Sonnet 5 $0.00011 $0.00332
Haiku 4.5 $0.00005 $0.00166

Measured yesterday against content hash 91bd3ab739eb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

peer-agents scanned grade B 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 yesterday.

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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

**Model defaults are frontier, always.** claude runs `--model fable` (alias tracking the latest Claude model — currently claude-fable-5 on a Max subscription); codex runs with no `-m`, so it uses the model from the host'
.agents/skills/peer-agents/SKILL.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

peer-agents

scripts/peer_agent.py runs claude -p or codex exec as a headless peer agent. The peer spends ITS OWN subscription budget (Claude Max / ChatGPT Pro), not your context — your only cost is launching the job and reading back the result file. Both CLIs must be installed and logged in (claude --version, codex login status).

How to dispatch

Launch as a background Bash task (peers take minutes, not seconds), then read the --out file when the completion notification arrives:

# Review this repo with Claude (read-only default):
python scripts/peer_agent.py claude --out output/peer-review.md \
    --prompt-file brief.md

# Have Codex fix something in a worktree (write mode):
python scripts/peer_agent.py codex --out output/codex-fix.md \
    --prompt "Fix the failing test in tests/test_universe_nodes.py and run it" \
    --cwd ../wf-bug126 --write

# Quick foreground question (prints to stdout, no file):
echo "One paragraph: what does workflow/router.py do?" | python scripts/peer_agent.py claude

For big briefs, write the brief to a file with your Write tool and pass --prompt-file — avoids shell-quoting and Windows command-line limits. The prompt always goes to the peer via stdin.

A dispatched peer must not dispatch

State this in every brief. A peer given a review brief will, left to itself, farm the review out rather than do it: on 2026-08-27 two dispatched Codex reviews each spawned their own peer_agent.py claude children (four in total), created three worktrees, and ran full local pytest -m "not slow" sweeps. After 34 minutes neither had written a single byte to its --out file, and both had to be killed. The work was recursive, not deep.

Put a constraints block in the brief itself -- the CLI has no flag for it:

HARD CONSTRAINTS ON HOW YOU WORK:
- Do NOT dispatch sub-agents. No scripts/peer_agent.py, no claude/codex
  subprocess, no new worktree. You are the reviewer; review it yourself.
- Do NOT run the full suite. No scripts/ci_required_tests.py, no
  `pytest -m "not slow"`. Run at most the one test file you need.
- Budget ~10 minutes. Read the diff and the cited files and reason.

Read the full file on GitHub · 97 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. yesterday First seen · 97 lines · 53 tokens per session scan B 91bd3ab739eb

Subscribe to this mod's changes

peer-agents is a skill published in the GitHub repository Jonnyton/TinyAssets (1 stars, last pushed 2d ago), licensed MIT. It adds 53 tokens to every session and 1,659 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens