Agent skill · Audited from source
carrel
skill · FengZhenfei/carrelMIT licenceAudited 12 Nov 2026 · source ad102dcb1ec1
Is it safe? With care — CAUTION, grade C (76/100). We read this skill's source at commit ad102dcb1ec1 and observed filesystem and network access and no hard-coded secrets. The audit raised 24 findings, 0 critical or high. What it does: Query the user's local knowledge bases through the Carrel search API, use original text and source images as needed in the answer, list the reference files and offer follow-up directions drawn from the knowledge graph.…
Cite
Cite this page
ModelsAtlas. “carrel (FengZhenfei/carrel)”. Retrieved 12 Nov 2026. https://modelsatlas.com/skills/fengzhenfei/carrel
Trust audit
Audit 0.4.1 · source read 12 Nov 2026
CAUTION means the audit found things worth knowing before you trust this skill. Grade C is shared by 105 of 18,389 audited skills.
What the source does
- FilesystemDeclared6 observations
- NetworkDeclared8 observations
- ShellNone observedNo commands run in the source
- DependenciesPinnedNothing floats
- SecretsNone foundNo keys in source
Five audit layers
| Layer | What we check | Result | Findings |
|---|---|---|---|
L0 Provenance | Repo owner, licence, archive status, activity | PASS | 0 |
L1 Static source | Secrets, path traversal, supply chain, hidden code | WARN | 24 |
L2 Instructions | Prompt injection, authority framing, hidden directives | PASS | 0 |
L3 Inventory | The files and declarations match the source | PASS | 0 |
L4 Sandbox | Run in isolation and watch behaviour | SKIPPED why | 0 |
Gates applied: no behavioural pass.
Findings (24)
MEDIUMInformation disclosuredisclose.log_secret · app/kb_pipeline/graph/schema_flow.py:369
print(f"…", flush=True)
- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMNetwork egressnet.raw_ip · app/kb_pipeline/config.py:160
opensearch_url=os.getenv("OPENSEARCH_URL", "http://127.0.0.1:9200"),- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMObfuscation / stealthobf.homoglyph · app/kb_pipeline/headings.py:65
_UNIT_AFTER_NUMBER_RE = re.compile(r"^[A-Za-zΩμ°%]{1,2}(?:\s|$)")- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMObfuscation / stealthobf.homoglyph · app/kb_pipeline/measure_units.py:22
"ma": "mA", "毫安": "mA", "milliamp": "mA", "milliamps": "mA", "ua": "μA", "μa": "μA", "μa": "μA", "微安": "μA",
- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMObfuscation / stealthobf.homoglyph · app/kb_pipeline/measure_units.py:24
"ns": "ns", "纳秒": "ns", "ps": "ps", "us": "μs", "μs": "μs", "μs": "μs", "微秒": "μs", "ms": "ms", "毫秒": "ms", "s": "s", "秒": "s",
- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMObfuscation / stealthobf.homoglyph · app/kb_pipeline/measure_units.py:27
"mw": "mW", "毫瓦": "mW", "w": "W", "瓦": "W", "ohm": "Ω", "ohms": "Ω", "ω": "Ω", "欧姆": "Ω", "kω": "kΩ", "kohm": "kΩ",
- Why
- No further detail recorded for this rule.
- Fix
- —
MEDIUMObfuscation / stealthobf.homoglyph · app/kb_pipeline/measure_units.py:28
"pf": "pF", "nf": "nF", "uf": "μF", "μf": "μF", "μf": "μF",
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWCode injectioncode.eval_exec · app/tests/test_parsers.py:1209
exec(compile(code, "fill.py", "exec"), ns)
- Why
- evaluates text as code
- Fix
- remove; use a parser or a dispatch table
LOWInsecure cryptocrypto.weak_hash · app/kb_pipeline/graph/compile.py:117
return hashlib.sha1(str(key or "").encode("utf-8")).hexdigest()[:8]- Why
- No further detail recorded for this rule.
- Fix
- —
LOWInsecure cryptocrypto.weak_hash · app/kb_server/api.py:208
digest = hashlib.sha1(json.dumps(rows, ensure_ascii=False, sort_keys=True, default=str).encode("utf-8")).hexdigest()- Why
- No further detail recorded for this rule.
- Fix
- —
LOWFilesystem / pathfs.traversal · app/tests/test_container_runtime.py:101
for invalid in [{"kb_id": "../../invalid"}, {"kb_id": "kb_001", "append": "yes"}]:- Why
- No further detail recorded for this rule.
- Fix
- —
LOWFilesystem / pathfs.traversal · app/tests/test_deployment.py:207
self.assertIn("../../models/mineru", mineru)- Why
- No further detail recorded for this rule.
- Fix
- —
LOWFilesystem / pathfs.traversal · app/tests/test_deployment.py:208
self.assertIn("../../runtime/mineru-output", mineru)- Why
- No further detail recorded for this rule.
- Fix
- —
LOWFilesystem / pathfs.traversal · app/tests/test_search.py:3166
for bad in ("../outside.png", "/etc/passwd", "kb/../../x.png", ""):- Why
- No further detail recorded for this rule.
- Fix
- —
LOWFilesystem / pathfs.traversal · app/tests/test_xlsx_images.py:81
@pytest.mark.parametrize('target,mode', [('https://example.invalid/image.png','External'),('../../../../image.png',''),('file:///secret.png','')])- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.raw_ip · README.md:135
Open the console at **http://127.0.0.1:9800**. Put documents in
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.raw_ip · README.md:149
curl -sS http://127.0.0.1:9810/search \
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.raw_ip · README.zh-CN.md:96
打开控制台 **http://127.0.0.1:9800**。将资料放入 `data/runtime/mirror/<目录名>/`,然后在控制台启用该目录。每个启用的目录对应一个知识库,文件由定时任务自动处理。
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWNetwork egressnet.raw_ip · README.zh-CN.md:105
curl -sS http://127.0.0.1:9810/search \
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWObfuscation / stealthobf.decode_call · app/kb_pipeline/parsers/mineru_pdf.py:201
raw = base64.b64decode(payload)
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWObfuscation / stealthobf.decode_call · app/kb_pipeline/parsers/notebook.py:141
image = base64.b64decode(re.sub(r"\s+", "", _text(data[mime])), validate=True)
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWObfuscation / stealthobf.decode_call · app/kb_search/api.py:186
data = base64.b64decode(raw, validate=False)
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWObfuscation / stealthobf.decode_call · app/tests/test_llm_multimodal_probe.py:99
with Image.open(io.BytesIO(base64.b64decode(data))) as picture:
- Why
- No further detail recorded for this rule.
- Fix
- —
LOWObfuscation / stealthobf.decode_call · app/tests/test_model_endpoints.py:407
with Image.open(BytesIO(b64decode(url.split(",", 1)[1]))) as image:- Why
- No further detail recorded for this rule.
- Fix
- —
Audit 0.4.1 · source sha ad102dcb1ec1 · 12 Nov 2026Report an issue or request a re-scan
Provenance
Who published it, and what the verdict is pinned to
- PublisherFengZhenfei GitHub userOwner of the repository this skill is read from
- Source folderFengZhenfei/carrelThe exact folder we audited
- LicenceMITAs the repository states it
- Commit pinned for this verdictad102dcb1ec1A new audit runs when the source changes
- Repository statusNot archivedLast push 3 days ago
- Sandboxed runNot runSkills are read, not executed; layer L4 is skipped
What it tells the agent
Full source on GitHub ↗Its description: Query the user's local knowledge bases through the Carrel search API, use original text and source images as needed in the answer, list the reference files and offer follow-up directions drawn from the knowledge graph.…
--- name: carrel-search description: "Query the user's local knowledge bases through the Carrel search API, use original text and source images as needed in the answer, list the reference files and offer follow-up directions drawn from the knowledge graph. Use when the user asks to search the knowledge base, find evidence in the stored material, compare records, or answer from several documents together." --- # Carrel-Search Evidence is retrieved through Carrel's search service (`app/kb_search`, default port 9810); the agent invoking this skill understands the question, composes the answer, lists the reference files and turns the graph leads into directions the user can ask about next. The service only recalls, reranks and arranges evidence and does no generation. The display name is **Carrel-Search** and the invocation name is `$carrel-search`. ## Invocation The client is `scripts/carrel_search.py` in this directory (Python 3.9+, standard library only). It defaults to `http://127.0.0.1:9810`; when the service runs on another machine, set the address and token through environment variables or a config file as described in [API and configuration](references/api.md). The skill contains no real token; never read or print one. `SKILL_DIR` stands for the directory containing this `SKILL.md`; replace it with the actual absolute path when executing. ```bash python3 "$SKILL_DIR/scripts/carrel_search.py" search --question "<the user's question>" ``` - **Every command prints a compact view directly**: retrieval status, the text of the sources (each with the file and place it comes from), facts, page summaries and graph leads, and at the end "Reference files", one line per file. Quality and degradation flags remain visible; IDs, scores and debug fields are left out; document text is not: a hit is printed in full and the chunks before and after it follow it. The script keeps the complete response by itself; do not pass `--output` and do not write scripts of your own to parse it. - **What does not fit goes to the next page; nothing is cut off**: one output has a size limit, and every page ends by saying what the next page is and how to fetch it (`show <call id> --page 2`). Where it says "read on", read every page before using the result: reading back the original and listings are always like this, and so is a search whose best sources are too long for the first page, which then says so. Where it says "the full text of the other sources, to be read when useful", the next page holds the lower-ranked sources and neighbouring chunks of a search and may be skipped. An entry too long for a page continues on the next one, which opens with the same label and "(continued from the previous page)". To save room, the path of a file is written in full only the first time and "same file as S1" afterwards, and a table header repeated from an earlier chunk reads "HEADER: same as S3". "chunk 26 of 67" at the end of a source line is its position in the document and the document's length in chunks. - **Every entry has a label**: S a source, F a fact, P a page, E an entity, R a relation, H the neighbourhood of a subject (`H1.2` is its second relation), N a relation returned by a neighbourhood lookup. Later commands point at an entry with `--ref <call id>:<label>` instead of copying IDs or writing request files; only R cannot be used this way: a relation has two ends and is no starting point, so point at the entity entry of one of its ends. The table gives the subcommand and arguments only; each follows the same script: | To do this | Subcommand | |---|---| | Read the original around a chunk | `context --ref a1b2c3:S3` (`--before` / `--after` set how many chunks on each side, 1 by default; read on with `--after` when the matter continues on the following pages). A fact (F) and a neighbourhood relation (N) can be read back the same way | | Read returned originals without summaries and graph leads | `show a1b2c3 --sources`; follow its numbered pages or read specific S references | | Verify document identity, publication time or version | `context --ref a1b2c3:S3 --opening` reads the first indexed chunk, preserving the document version and time scope | | Read a whole document (a whole table, a whole report) | `context --ref a1b2c3:S3 --whole`: one command, printed page by page; a document of more than 300 chunks is read in parts with `--before` / `--after` | | See the original picture, crop a part | `image --ref a1b2c3:S6`; `crop --ref a1b2c3:S6 --bbox 0.1,0.2,0.6,0.8`. View the saved file. `crop` only cuts a part out; it is no sharper than the original | | Walk one hop along the graph from an entity | `neighbors --ref a1b2c3:H1` (E, `H1.2` and N work the same; `--type` takes one kind of relation only, `--limit` up to 100) | | List the facts under an entity | `facts --ref a1b2c3:H1` (add `--property name` for one property) | | Search again inside one document | `search --question "..." --in-doc a1b2c3:S3` (S, D, G and F references work; the knowledge base follows the document; with `--kb` given and the document outside it, the command fails) | | Look back in time | `search`, `grep`, `neighbors` and `facts` accept `--as-of 2024` (year/month forms use the last day) or `--history` (include superseded, expired and archived records with status flags) | | Inspect types and coverage | `catalog` includes graph overviews; use `schema --kb <KB id>` for type inheritance and relation definitions, and `stats --kb <KB id>` for coverage and time spans | | See the full text or every field of an entry | `show a1b2c3:S9 a1b2c3:S13` (full text for sources and pages, every field for the rest); `show a1b2c3` prints the compact view of that call again and `show a1b2c3 --page 2` its next page | - Limit the scope with arguments: `--kb` (repeatable), `--top-k`, `--block-type table` (a soft preference for table lookups). Do not hard-code KB numbers and do not assume every knowledge base has a graph. Send commands th
Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.
Install
Pinned to the audited commit ad102dc
git clone https://github.com/FengZhenfei/carrel.git
git clone https://github.com/FengZhenfei/carrel git -C carrel checkout ad102dcb1ec1
Commands as the repository documents them: shown, never run by us. Our verdict covers commit ad102dcb1ec1 only.
Works in which agents?
What its own text says, not a test
| Agent | Status | More |
|---|---|---|
| Claude Code | Mentioned in its text, not tested | Skills for Claude Code · About Claude Code |
| Codex | Mentioned in its text, not tested | Skills for Codex · About Codex |
| Cursor | Mentioned in its text, not tested | Skills for Cursor · About Cursor |
Audit history
Every audit, every source change
- !CAUTION C 76 / 100
first audit · source
ad102dcb1ec1
README badge
For maintainers · shows the latest audit
[](https://modelsatlas.com/skills/fengzhenfei/carrel/)
The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.
Similar skills
All data & analysis skills →tirth8205/code-review-graph
Build or update the code review knowledge graph. Run this first to initialize, or let hooks keep it updated automatically.
tirth8205/code-review-graph
Systematically debug issues using graph-powered code navigation
tirth8205/code-review-graph
Navigate and understand codebase structure using the knowledge graph
tirth8205/code-review-graph
Plan and execute safe refactoring using dependency analysis
tirth8205/code-review-graph
Perform a structured code review using change detection and impact
tirth8205/code-review-graph
Review only changes since last commit using impact analysis. Token-efficient delta review with automatic blast-radius detection.
Add this badge to your README
[](https://modelsatlas.com/skills/fengzhenfei/carrel/)
<a href="https://modelsatlas.com/skills/fengzhenfei/carrel/"><img src="https://modelsatlas.com/skills/fengzhenfei/carrel/badge.svg" alt="ModelsAtlas: CAUTION C"></a>
Questions about carrel
What does the carrel skill do?
Is carrel safe to install?
What can carrel access on my machine?
Which assistants does carrel work with?
How current is this page?
Sources
- github.com/FengZhenfei/carrel — the audited source at commit ad102dcb1ec1 and every finding aboveObserved · 12 Nov 2026
Spotted something wrong? Report a correction · Methodology
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- What does the carrel skill do?
- Is carrel safe to install?
- What can carrel access on my machine?