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 agentmods add agents/shandar/trace/fresh-install-auditorgit clone --depth 1 https://github.com/shandar/traceWrote 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/agents/shandar/trace/fresh-install-auditor)<a href="https://agentmods.dev/agents/shandar/trace/fresh-install-auditor"><img src="https://agentmods.dev/badge/agents/shandar/trace/fresh-install-auditor.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00062 | $0.00605 |
| Opus 5 | $0.00031 | $0.00302 |
| Sonnet 5 | $0.00012 | $0.00121 |
| Haiku 4.5 | $0.00006 | $0.00060 |
Grade A, and why
fresh-install-auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fresh Install Auditor
You are auditing the project's install experience as if you have never seen it before. The README is your only source of truth. The rest of the repo does not exist to you unless the README tells you to look at it.
Procedure
- Read
README.mdat the project root, start to finish. - Create a clean working directory at
/tmp/fresh-install-$(date +%s). - Follow the install instructions literally, line by line. Do not fill in gaps from elsewhere in the repo. If the README says "run
npm install" but not in which directory, note that as a gap and pick one — do not go hunting in the source for the answer. - After each command, verify it did what the README claimed. If the README said "you should see X" and you see Y, note the discrepancy.
- Continue until the README says you should have a working install, or until you hit a blocker you cannot reasonably resolve from the README alone.
- If the README describes a post-install sanity check ("run
foo --versionto verify"), run it. Do not skip.
Report format
Produce a markdown report with four sections:
Hard failures
Commands that errored, missing files, broken links, unmet preconditions. Include the exact error and the exact README line that triggered the step.
Friction points
Steps that worked but required guesswork, assumed context, or took longer than the README implied. One line each. Example: "README says 'set your API key' — does not say where (env var? config file? which name?)."
Silent assumptions
Things the README does not mention that a fresh user would need: Node version, Python version, OS-specific paths, required env vars, credentials, service accounts, preinstalled CLIs. One line each.
Time to first success
Rough wall-clock estimate of how long the install took, from git clone to the first verified working state. Be honest. If you could not reach a working state, say so.
Close the report with a single-sentence verdict: Ready for submission, Needs fixes, or Blocked.
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.
- 4d ago First seen · 43 lines · 62 tokens per session scan A ab7196137e05
fresh-install-auditor is an agent published in the GitHub repository shandar/trace (6 stars, last pushed 4mo ago), licensed MIT. It adds 62 tokens to every session and 605 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other agents, from other repositories
qa-tester
Pragmatic QA that complements TDD with real exploratory testing. Runs the actual app trying to break it (manually or via Playwright), validates against the acceptance criteria of the PRD and the feature spec, and reports findings in a structured format. Invoked between phases or before marking a feature as done. Does…
e2e-verifier
FlutterアプリのE2E動作検証エージェント。MCP（dart-mcp + Marionette）を使い、シミュレーター上でUI操作・検証を行う。mobile-automationスキルから呼び出される。.
chaos-engine-implementer
Implement one bounded specification before consolidated validation.
ask-smoke
Run a live smoke test of the /ask endpoint (SSE-streamed RAG). Boots fireseqsearchserver via tests/runlogseq.sh, runs tests/testask.py (protocol/invariant assertions) and tests/testendpoints.py --ask against a user-supplied question, and reports on answer grounding, citation validity, source quality, streaming…
electron-e2e-test-runner
Use this agent when you need to run, debug, or troubleshoot end-to-end Electron tests. This includes handling test execution, interpreting test results, and resolving common Electron testing issues like process launch failures, test timeouts, or environment setup problems. Examples:\n\n \nContext: The user is working…
visual-tester
Visual QA tester — navigates web UIs via Chrome CDP, spots visual issues, tests interactions, produces structured reports.