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 Kastalien-Research/thoughtbox --skill experimentgit clone --depth 1 https://github.com/Kastalien-Research/thoughtboxWrote 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/kastalien-research/thoughtbox/experiment)<a href="https://agentmods.dev/skills/kastalien-research/thoughtbox/experiment"><img src="https://agentmods.dev/badge/skills/kastalien-research/thoughtbox/experiment/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/kastalien-research/thoughtbox/experiment"><img src="https://agentmods.dev/badge/skills/kastalien-research/thoughtbox/experiment.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 | $0.00053 | $0.01564 |
| Opus 5.5 | $0.00021 | $0.00626 |
| Sonnet 5.5 | $0.00011 | $0.00313 |
| Haiku 4.5 | $0.00005 | $0.00156 |
Grade A, and why
experiment 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 1mo 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 — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run a parallel experiment: $ARGUMENTS
Purpose
Test multiple competing approaches to the same problem simultaneously. Each approach gets its own git worktree and sub-agent. Results are compared at the end so the best approach can be merged.
This is NOT for parallelizing parts of one plan (use /workflows-work for that). This is for when you genuinely don't know which approach is best and want to try all of them.
Process
Phase 1: Define Tracks (orchestrator does this)
Parse the input (a gap analysis, a list of alternatives, a brainstorm, or a user description) and produce a track list:
TRACK DEFINITION
================
Base branch: <current branch or specified base>
Track A: <name>
Branch: exp/<slug-a>
Approach: <1-2 sentence description>
Files touched: <predicted file list>
Effort: low | medium | high
Track B: <name>
Branch: exp/<slug-b>
...
[up to 5 tracks]
Conflict check: If two tracks modify the same files in incompatible ways, flag this to the user before proceeding. Parallel experiments must be independent.
Present the track list to the user and wait for confirmation before proceeding.
Phase 2: Create Worktrees
For each track, create a worktree branching from the base:
# From the repo root
git worktree add ../<repo-name>-exp-<slug> -b exp/<slug>
Worktree naming convention: ../<repo-name>-exp-<slug> (sibling to the main repo directory).
Verify all worktrees were created:
git worktree list
Phase 3: Dispatch Sub-Agents (parallel)
Dispatch one sub-agent per track using the Agent tool. All independent tracks run in parallel.
Each sub-agent receives:
- Working directory: The worktree path (CRITICAL — sub-agent must
cdhere) - Track definition: What to implement and why
- Source context: Relevant file contents from the base branch (read before dispatching)
- Acceptance criteria: What "done" looks like for this track
- Instructions:
- Work ONLY in your assigned worktree directory
- Write tests for your changes
- Run tests and report results
- Commit your work to the worktree's branch
- Return the structured summary (see format below)
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.
- 1mo ago First seen · 199 lines · 53 tokens per session scan A 9fa17324ce72
experiment is a skill published in the GitHub repository Kastalien-Research/thoughtbox (65 stars, last pushed 3mo ago), licensed MIT. It adds 53 tokens to every session and 1,564 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
datalad
Retrieves, versions, and publishes scientific datasets with DataLad and git-annex, and captures computational provenance with datalad run, rerun, and containers-run. Use when cloning or fetching data from OpenNeuro, DANDI, datasets.datalad.org, or any DataLad dataset; when a file in a dataset reads as a broken symlink…
commit-context
Trace a file, function, or line back to the agent session that produced its current commit. Use when the user asks "why is this code here", "what was the agent doing when this changed", "who wrote this", or wants context on a specific location in the codebase.
commit-history
List recent git commits linked to agent sessions, optionally filtered by branch or repo. Use when the user asks "show agent commits", "what has the agent shipped", "list linked commits", or wants commits with their session context.
baby-sit
Monitor a GitHub pull request until CI is green, diagnose failures, and rerun only evidence-backed flaky GitHub Actions jobs.
cyrus-setup-repository
Add one or more Git repositories to Cyrus configuration so it can process issues from those repos.
github-commenting
How to post clean, rich, deduplicated GitHub PR review comments — suggestion blocks, multi-line anchors, markers, formatting rules. Load before posting or fixing any PR comment.