Skip to content
Back to search

agent-team-skills

MarketplaceWorks with Claude CodeGitHub

Skills for leading a team of Claude Code agents without paying for them to remember.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/coracle-us--agent-team-skills/badge.svg)](/listings/coracle-us--agent-team-skills)

HTML

<a href="/listings/coracle-us--agent-team-skills"><img src="/api/listings/coracle-us--agent-team-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

In Claude Code:

/plugin install agent-team-skills --marketplace coracle-us/agent-team-skills

On older versions, two steps:

/plugin marketplace add coracle-us/agent-team-skills
/plugin install agent-team-skills@agent-team-skills

Then check your own week: ask Claude "where did my tokens go this week?" and the `audit-your-spend` skill runs.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 17 lines
  1. {
  2. "name": "agent-team-skills",
  3. "description": "Skills for leading a team of Claude Code agents without paying for them to remember.",
  4. "owner": {
  5. "name": "Jon McBee (Coracle)",
  6. "email": "jon@coracle.us"
  7. },
  8. "plugins": [
  9. {
  10. "name": "agent-team-skills",
  11. "description": "Skills for leading a team of Claude Code agents without paying for them to remember.",
  12. "version": "0.1.0",
  13. "source": "./",
  14. "author": { "name": "Jon McBee", "email": "jon@coracle.us" }
  15. }
  16. ]
  17. }
README.md · README · 97 lines
  1. # agent-team-skills
  2. **In agentic coding you pay for agents to remember, not to work.**
  3. In October 2026 I used 25.5 billion tokens through Claude Code, most of it leading teams of agents that built a multiplayer
  4. browser game in 12 days. Only **0.24%** of those tokens were the model writing anything: code, prose, commands.
  5. **96.4%** was the model re-reading context it already had, from the cache, on every single call.
  6. So I audited every transcript to find where the re-reading went, and turned what I found into seven Claude Code
  7. skills. Each one fires at the moment the leak happens, for whoever is **leading a team of agents**.
  8. ## Install
  9. In Claude Code:
  10. ```text
  11. /plugin install agent-team-skills --marketplace coracle-us/agent-team-skills
  12. ```
  13. On older versions, two steps:
  14. ```text
  15. /plugin marketplace add coracle-us/agent-team-skills
  16. /plugin install agent-team-skills@agent-team-skills
  17. ```
  18. Then check your own week: ask Claude "where did my tokens go this week?" and the `audit-your-spend` skill runs.
  19. ## What I found, and the skill for each
  20. From the 12-day build (Oct 18–30: 21.9B tokens, 292 agents plus 2 lead sessions), as `audit-your-spend` counts it. Cost shares are estimates from API
  21. list-price ratios.
  22. | Skill | Fires when | The leak it stops |
  23. |---|---|---|
  24. | `briefing-agents` | you're about to spawn an agent | Agents were **86%** of tokens. One agent kept alive for **117 hours** made 1,320 calls at ~500k context each: **12%** of all cost by itself. |
  25. | `waiting-for-free` | a build, CI run or deploy will take minutes | over **2,000** sleep-and-check loops, and **1,154** cache rebuilds after idle gaps: about **35%** of all cost. |
  26. | `context-hygiene` | a session is getting heavy, or CLAUDE.md is growing | Calls carrying over 400k context were **46%** of calls and **70%** of cost. CLAUDE.md grew from 10 KB to 255 KB, and a new agent's starting context went from **32k to 113k** tokens. |
  27. | `looking-cheaply` | you're about to take screenshots | More than **6,500** screenshots, each re-read on every later turn. |
  28. | `merging-in-batches` | you're merging a stack of PRs, or a test is flaky | One CI run for a resolved stack, not one per PR; a flaky test gets one logged retry, not a restart. |
  29. | `rulings-become-tests` | the lead makes a call that must keep holding | Fresh agents don't know yesterday's decisions. A test does. |
  30. | `audit-your-spend` | you ask where the tokens went | Runs locally on your own logs and shows all of the above for your week. |
  31. ### `briefing-agents`
  32. Agents are disposable: one job, one result, then the agent ends. The skill gives the shape of a brief (goal, where to
  33. look, what's ruled out, done when) and a footer to paste into every brief: stop your servers, never end a turn
  34. waiting on a job, no sleep loops, state goes in files.
  35. ### `waiting-for-free`
  36. Waiting should cost nothing. Let the platform wait (`gh pr merge --auto`), or run one background waiter that wakes the
  37. agent once. Includes the three waiter traps that bit us: `pgrep -f` matching its own wait loop, a "done" grep firing
  38. on the previous run's log line, and completion notices that don't wake an idle agent.
  39. ### `context-hygiene`
  40. You pay for the window you carry, not the window you have. Near 250k tokens, write a handoff file and start fresh.
  41. Keep CLAUDE.md an index under about 20 KB; detail goes in docs agents read only when their task touches it.
  42. ### `looking-cheaply`
  43. Capture everything, pixel-diff against the last run, view one labelled contact sheet, and zoom only where something
  44. changed. Or hand the looking to a short-lived reviewer that reports in words.
  45. ### `merging-in-batches`
  46. Run the slow check once on the tip of a resolved stack, not once per PR. A flaky test gets one logged retry; two
  47. failures stop the queue; then fix the root (pin the clock and the seed).
  48. ### `rulings-become-tests`
  49. When the lead says "note that" about a rule, write the guard test in the same turn, with the ruling quoted in its
  50. failure message, and one line in CLAUDE.md pointing at it.
  51. ### `audit-your-spend`
  52. A standard-library Python script that reads `~/.claude/projects` and prints counts and percentages: tokens by day and
  53. week, the mix, heavy calls, cache rebuilds, the costliest sessions and how long they lived, starting context by day,
  54. polls and images. It's read-only, makes no network calls, and never prints message text, project names or prices.
  55. Run it directly from a clone of this repo:
  56. ```bash
  57. python3 skills/audit-your-spend/scripts/audit.py --since 2026-11-01 --until 2026-11-07
  58. ```
  59. ## How the skills were tested
  60. Each behaviour skill has a scenario in `tests/scenarios/`: a fresh agent answers it once without the skill and once
  61. with it (`zsh tests/run-scenarios.sh <out-dir> [skills-root]`). Without the skills, a fresh agent views all 12 screenshots
  62. at full size, keeps working at 300k context ("300k of 1M is not a problem"), and runs CI five times on a stack it
  63. could check once. With them, it doesn't. The audit script and the repo have unit tests
  64. (`python3 -m unittest discover -s tests`), and on my own logs the audit reproduces my October total to the token.
  65. ## Licence
  66. MIT. By Jon McBee, [Coracle](https://github.com/coracle-us).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.