Borrowing it
Nothing to install: this file belongs to ashaychangwani/imprint. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ashaychangwani/imprint/main/.agents/skills/imprint-reteach-audit/SKILL.mdgit clone --depth 1 https://github.com/ashaychangwani/imprintWrote 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/ashaychangwani/imprint/imprint-reteach-audit)<a href="https://agentmods.dev/skills/ashaychangwani/imprint/imprint-reteach-audit"><img src="https://agentmods.dev/badge/skills/ashaychangwani/imprint/imprint-reteach-audit/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/ashaychangwani/imprint/imprint-reteach-audit"><img src="https://agentmods.dev/badge/skills/ashaychangwani/imprint/imprint-reteach-audit.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 | $0.00033 | $0.01553 |
| Opus 5.5 | $0.00013 | $0.00621 |
| Sonnet 5.5 | $0.00007 | $0.00311 |
| Haiku 4.5 | $0.00003 | $0.00155 |
Grade A, and why
imprint-reteach-audit 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 15d 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Re-teach + audit imprint sites
Use this to rebuild a site's tools from its existing recording and confirm they work.
Fresh master-flow validation
Check the branch and timeline.md first. For continuing the September 2026
experiment, read the handoff. Its local
paths and process state are historical facts to verify, not instructions to
resume old processes or assumptions that the data exists on another machine.
Use a new isolated IMPRINT_HOME and an explicit recording. Do not use a
cleanup script that erases earlier output during a repeatability experiment.
After code or prompt changes, start fresh, never resume a failed teach.
Do not feed shipped examples or prior generated artifacts to the teaching agent.
Commands
# Replace placeholders with verified paths; do not reuse a prior run home.
IMPRINT_HOME=<new-isolated-home> bun run src/cli.ts teach <site> \
--agent codex --from-session <recording.json> --timeout 90m --no-interactive
# Audit (spawns a headless Codex that calls every compiled tool live and grades
# correct/broken; exit 0 = PASS, default threshold 95%):
IMPRINT_HOME=<same-isolated-home> bun run src/cli.ts audit <site> \
--provider codex-cli --timeout 45m --json
Hard rules
- Codex fresh teaches use a native family capped at ten concurrent children. Preserve native assignment, lifecycle and descendant usage history when comparing runs. Other providers retain four focused workers. Run validation teaches and audits sequentially.
- Respect the user's deadline. In this experiment the target is about 30 minutes, a reasoned progress decision at 60 minutes, and a 90-minute hard deadline. These are not universal defaults or permission to extend a run silently.
- Use the product's monitoring mechanism for long runs; stop task-owned monitors at handoff. A quiet agent is not by itself stuck. Check current facts before killing a run. No fixed repair-attempt limit is implied.
- To re-teach several sites + audit each as one hands-off job, loop them sequentially in a single background driver script (teach → audit → next).
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.
- 15d ago Changed · +44 lines · -11 tokens per session 26a9f23cf6a4
- 2mo ago First seen · 60 lines · 44 tokens per session scan A 783865d46781
imprint-reteach-audit is a skill published in the GitHub repository ashaychangwani/imprint (24 stars, last pushed 15d ago), licensed MIT. It adds 33 tokens to every session and 1,553 once invoked, about $0.0001 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
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers hosted cloud browsers for AI agents: the browser-use Python library (Browser(usecloud=True)), Playwright, Puppeteer or any CDP client.
x402
Set up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key. Two setups it works out up front — "just use it" (set up a wallet so you or Claude Code can run cloud browser tasks paid from the wallet — Claude writes and runs throwaway scripts, nothing…
tool-prompt-optimization
Optimize the description prompts an AI agent reads to learn its built-in tools (the .md files under prompts/tools/). Two halves: (1) measure how much of a prompt is already inferable from the tool's JSON parameter schema + name, to prune redundancy with evidence; (2) house authoring rules for what belongs in a tool…
open-source
Documentation reference for writing Python code using the browser-use open-source library. Use this skill whenever the user needs help with Agent, Browser, or Tools configuration, is writing code that imports from browseruse, asks about @sandbox deployment, supported LLM models, Actor API, custom tools, lifecycle…
clinical-reports
Write comprehensive clinical reports including case reports (CARE guidelines), diagnostic reports (radiology/pathology/lab), clinical trial reports (ICH-E3, SAE, CSR), and patient documentation (SOAP, H&P, discharge summaries). Full support with templates, regulatory compliance (HIPAA, FDA, ICH-GCP), and validation…
imaging-data-commons
Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index. Use for accessing large-scale radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required. Query by metadata, visualize in browser, check licenses.