보안약점 점검 (개발보안 가이드)

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With readOnlyHint=false, the description explains why: although projectDir is read-only, the tool can write SARIF output (sarifPath, overwrite). It also discloses non-obvious behavior the annotations cannot — rule source, comment-skipping, output fields (file/line/CWE/severity/remedy plus provenance), baseline diffing, SARIF 2.1.0 for GitHub code scanning, and the authoritative-diagnosis caveat.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose and rule source are front-loaded in the first clause, and every later sentence carries actionable information. The long enumeration of vulnerability classes is informative but heavy, and the single dense paragraph makes the description hard to skim; slightly tighter grouping of output/exception/baseline behavior would help.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool is complex (10 params, 1 enum, output schema present) yet the description covers rule provenance, ignore-file prerequisites, baseline comparison, output formats, and the limits of static analysis, so an agent can call it and interpret results without opening the schema. The output schema handles return-value detail.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the schema already documents all ten parameters. The description still adds value by tying baselinePath to "new findings only", explaining format=sarif / sarifPath as SARIF 2.1.0 versus the maxFindings "all findings in file" distinction, and describing the exception-file contract — semantics beyond the schema text.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a precise verb and resource — static pattern-based search for implementation-stage security weaknesses in source, keyed to the MOIS/KISA 2021 guide — and enumerates the exact rule classes (SQL injection, XSS, path traversal, XXE, weak crypto, etc.) it covers. No sibling tool performs secure-coding analysis (the nearest siblings are SBOM/dependency checks), so the domain alone makes it unmistakable.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear operating context: rules come from catalog/secure-coding-rules.json, comments are ignored, exceptions go in .egovframe-secure-ignore.json with mandatory reason/reviewer, and baselinePath shows only newly introduced findings. It also scopes the tool honestly ("패턴 기반 진단 보조이며 공인 진단을 대체하지 않습니다"). It never explicitly says when to prefer another tool or which conditions make it inappropriate, so it stops short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.