owasp（owasp 搜索）

OWASP Top 10 Dispatcher

Dispatch parallel subagents covering all 10 categories of the OWASP Top 10 (2021). Each category runs as an independent subagent with its own context window, analyzing the scoped code for that specific class of vulnerability. Results are consolidated, deduplicated, and ranked by severity.

Supported Flags

Read ../../shared/schemas/flags.md for the full flag specification. This dispatcher supports all cross-cutting flags.

Framework Reference

Read ../../shared/frameworks/owasp-top10-2021.md for the full OWASP Top 10 specification including vulnerability descriptions, prevention guidance, and STRIDE cross-mappings for each category.

Pre-flight Relevance Check

Before dispatching subagents, scan the scoped file list to determine which categories are relevant. Skip categories that have no plausible attack surface in the codebase. This avoids wasting subagent context windows on irrelevant analysis.

How to check: Use Glob and Grep on the scoped files to detect relevant patterns. For example, check for fetch(, axios, requests.get, http.Get to determine A10 relevance. Check for package.json, requirements.txt, go.mod to determine A06 relevance.

If --only is specified, skip the relevance check and dispatch only the listed categories.

Dispatch Category Subagents

CRITICAL: All Task tool calls MUST appear in the SAME response message. This is what triggers parallel execution. If you emit them across separate messages, they run sequentially and waste time.

Dispatch Table

Subagent Prompt Template

Each subagent Task call must include a FULLY self-contained prompt. Subagents get their own isolated context window and cannot see the main conversation.

Each subagent prompt must contain:

1. The concrete file list to analyze (resolved from scope).
2. The absolute path to the category SKILL.md to read and follow.
3. The flags to apply (--scope, --depth, --severity, --format, etc.).
4. The findings schema path (shared/schemas/findings.md) for output format.
5. An instruction to return findings only -- no summary, no cross-category commentary. The dispatcher handles consolidation.

Analyze the following files for OWASP {CATEGORY_CODE} ({CATEGORY_NAME}) vulnerabilities:

FILES:
{FILE_LIST}

STEP 1: Read the skill definition at:
{ABSOLUTE_PATH_TO_PLUGIN}/skills/{SKILL_NAME}/SKILL.md

STEP 2: Follow the workflow defined in that skill to analyze the listed files.

STEP 3: Read the findings schema at:
{ABSOLUTE_PATH_TO_PLUGIN}/shared/schemas/findings.md

STEP 4: Output findings in the schema format. Set metadata.framework to "owasp"
and metadata.category to "{CATEGORY_CODE}".

FLAGS: --scope {SCOPE} --depth {DEPTH} --severity {SEVERITY}

IMPORTANT: Return ONLY the findings list. Do NOT produce a summary or
cross-category analysis. The dispatcher handles consolidation.

Launching

Emit one Task tool call per relevant category, ALL in a single response:

• subagent_type: "general-purpose"
• description: "OWASP {CATEGORY_CODE} - {CATEGORY_NAME}"
• prompt: The fully self-contained prompt above, filled in for this category.

Do NOT emit Task calls one at a time. Do NOT wait between dispatches.

Consolidation

After ALL subagents return their results:

1. Merge Findings

Collect all findings from all subagent responses into a single list.

2. Deduplicate

Two findings are duplicates if they share the same location.file AND location.line (or overlapping line ranges). When duplicates exist:

• Keep the finding with the higher severity.
• Merge cross-references (a finding tagged A03 and A01 keeps both).
• Note the duplicate in the retained finding's description.

3. Cross-Reference

For each finding, populate cross-framework references where known:

• references.cwe: Map to CWE identifier.
• references.stride: Map to STRIDE category letter(s).
• references.mitre_attck: Map to ATT&CK technique ID.
• references.sans_cwe25: Map to SANS/CWE Top 25 rank if applicable.

4. Rank by Severity

Sort findings: critical > high > medium > low. Within the same severity, sort by confidence (high > medium > low).

5. Produce Aggregate Output

Wrap the consolidated findings in the aggregate output format from shared/schemas/findings.md, including categories_checked, categories_skipped, skip_reason, total_findings, and by_severity.

6. Present Results

Output the consolidated report in the requested --format. Include:

• Summary table: categories checked, categories skipped (with reasons).
• Severity breakdown: count by critical/high/medium/low.
• Findings list in severity order.
• Cross-category patterns (e.g., "injection findings in A03 also indicate broken access control in A01").

Expert Mode

If --depth expert is set, after consolidation, launch red team subagents to simulate exploitation of the findings. Each red team agent receives the consolidated findings and attempts to construct multi-step attack chains.

Read ../../shared/frameworks/dread.md for DREAD scoring criteria. Each finding receives a DREAD score in expert mode.

Launch red team agents as parallel Task calls (same single-response rule). Pass the full findings list to each persona. Red team output is appended to findings with prefix RT and metadata.tool set to "red-team".