ir-search
🇰🇷 한국 정부 지원사업 전수조사 에이전트 스킬 [Claude Code·Codex·agy(Antigravity)·Cursor·Gemini CLI·Grok Build 지원] K-Startup·기업마당·NIPA·KOCCA·SMTECH 크롤링 + 즉시/로드맵/변형 3분류 (Korea-only)
Runssource: GitHubPythonMITcommit 9ef913a40f3a
Python, MIT licensed. The project labels itself: agent skills, antigravity, claude code, claude skill, codex, crawler, cursor and gemini cli.
ir-search runs. An Argusic agent installed it in 1 minute and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
The project's 229 pytest tests pass, and every CLI crawler (K-Startup, bizinfo, NIPA, KOCCA, SMTECH list and detail, scope_plan, diff_surveys) ran successfully against live Korean government sites, producing real announcement records and detail text.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.

What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- pip3 install 'curl_cffi>=0.15' failed with 'externally-managed-environment' (PEP 668) on the container's system Python; install.sh hit the same non-fatal error1 minute
- Install time
- 1 minutes
- Cold machine to finish
- 4 minutes
- Errors hit and fixed
- 1 hit, 1 fixed with no human help
- How the result was proved
- `.venv/bin/python -m pytest tests/ -q` exited 0 with 229 passed, 1 skipped; live crawls: kstartup_crawl.py list --max-pages 1 --smoke returned 15 items, sources_crawl.py list bizinfo/nipa/kocca/smtech returned 15/10/6/9, list all returned 40, kstartup_crawl.py detail 179577 and sources_crawl.py detail <bizinfo URL> saved detail text, scope_plan.py wrote scope-plan.json, diff_surveys.py classified NEW/CHANGED/CLOSED/UNCHANGED
- Model tokens used
- 59,610
- Exact commit tested
- 9ef913a40f3a
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
Let me finalize.I have completed a thorough verification. Let me compile the findings into the fina… **Summary of what I verified:** - **Install**: Python environment is PEP 668 externally-managed, so `pip3 install 'curl_cffi>=0.15'… - **Tests**: `pytest tests/` → **229 passed, 1 skipped** (the skip is an opt-in live API canary). - **Live end-to-end (tier1_real)** against real Korean government sites: - `kstartup_crawl.py list` → 15 items (curl_cffi backend) and 15 items via urllib fallback (syste… - `kstartup_crawl.py detail 179577` → detail text saved - `sources_crawl.py list bizinfo|nipa|kocca|smtech` → 15/10/6/9 items - `sources_crawl.py list all` → 40 items - `sources_crawl.py detail <bizinfo URL>` → detail text saved - `scope_plan.py --preset recommended` → scope-plan.json - `diff_surveys.py` → correctly classified NEW/CHANGED/CLOSED/UNCHANGED
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 1 minute, faster than the median of the 127 comparable projects Argusic has measured.
- Recovered from all 1 error without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 128 agent-skills projects Argusic has installed and timed, ir-search was the 35th fastest to reach a running state, and 96 of 128 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 1 minutes after the clone.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
agent-skillsantigravityclaude-codeclaude-skillcodexcrawlercursorgemini-cligovernment-grantsgrokkoreastartup
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/ir-search)Questions
- Does ir-search run?
- Yes. ir-search runs. Argusic installed and launched it on a clean machine in 1 minutes, hitting 1 error on the way, and recorded the session.
- How did Argusic test ir-search?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 9ef913a40f3a. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test ir-search?
- The run that produced this verdict cost $0.03: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does ir-search take to install?
- 1 minute in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does ir-search need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing ir-search?
- 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does ir-search compare with the alternatives?
- Of the 128 agent-skills projects Argusic has installed and timed, ir-search was the 35th fastest to reach a running state, and 96 of 128 reached one at all.
- Where is the evidence for ir-search?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.