Back to skill

Security audit

book-capture-obsidian

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this book-capture skill benign with high confidence: the scanners found no AIG issues, and the reported concerns are mostly context-dependent or related to setup behavior.

The verdict is benign, but benign does not mean risk-free. Before installing, consider that setup may write package caches outside .venv and the skill's tool scope is not declared; the scanner did not verify runtime permission limits.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (9)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_read, file_write, network, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/extract_isbn.py (reported line 30)May include surrounding context.

python
warnings: List[str] = []
    values: List[str] = []
    try:
        run = subprocess.run(
            [binary, "--quiet", image_path],
            check=False,
            capture_output=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/extract_isbn.py (reported line 127)May include surrounding context.

python
for config in configurations:
        try:
            run = subprocess.run(
                [binary, image_path, "stdout", "-l", lang, *config],
                check=False,
                capture_output=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · tests/test_ci_gate.py (reported line 19)May include surrounding context.

python
copied_ci = isolated_scripts / "run_ci_local.sh"
    shutil.copyfile(repository_root / "scripts" / "run_ci_local.sh", copied_ci)

    run = subprocess.run(
        ["sh", str(copied_ci)],
        check=False,
        capture_output=True,

Unbounded Resource Access

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.

Content

Scanner excerpt · tests/test_google_books.py (reported line 61)May include surrounding context.

python
requested_isbn=ISBN,
        api_key=SENTINEL,
        request_get=fake_get,
        max_retries=0,
    )

    assert captured["url"] == google_books.GOOGLE_BOOKS_URL

Unbounded Resource Access

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.

Content

Scanner excerpt · tests/test_metadata_resolver.py (reported line 159)May include surrounding context.

python
assert next_run.stats.google_skipped_cooldown == 1

    clock.t += 3601
    after_cooldown = make_resolver(fake_get, clock, MetadataCache(path, now=clock.now), google_max_retries=0)
    after_cooldown.lookup(isbn13="9780131103627")
    assert len([url for url in log if "googleapis" in url]) == 3

Unbounded Resource Access

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.

Content

Scanner excerpt · tests/test_metadata_resolver.py (reported line 375)May include surrounding context.

python
calls.append(url)
        assert "googleapis.com" in url
        return FakeResponse(status, payload)
    resolver = make_resolver(get, clock, providers=["google_books"], google_max_retries=0)
    resolver.lookup(isbn13=ISBN)
    resolver.lookup(isbn13="9780201633610")
    assert calls and resolver.stats.network_calls["openlibrary"] == 0

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · LICENSE (reported line 12)May include surrounding context.

text
permit persons to whom the Software is furnished to do so.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE

Setup claims filesystem changes are confined to .venv, but pip can write outside it.

Low
Category
Excessive Agency
Confidence
93% confidence
Finding

The comment at L03 explicitly promises that setup only touches /.venv. The pip commands at L18–L19 use pip's default caching behavior, which normally writes HTTP and package caches to a user-level cache directory outside that boundary; using a virtual environment does not isolate that cache.

Content

No source excerpt is available for this finding.