Skip to main content
Glama
delimit-ai

delimit

Official
by delimit-ai
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Delimit Security Audit

delimit_security_audit

Audit code for security vulnerabilities, hardcoded secrets, and dangerous patterns before release. Automatically creates evidence, governance tasks, and alerts on critical findings.

Instructions

Audit security and auto-chain evidence + governance on critical findings.

When to use: as the deploy gate / pre-release security check — combines dependency vulnerability scanning, hardcoded-secret detection, dangerous-pattern checks, and .env-tracked-in-git checks, AND automatically opens a governance task + sends a notification when critical findings are present. When NOT to use: for a baseline scanner pass without auto-chained side effects (use delimit_security_scan), to ingest an external scanner's output (delimit_security_ingest), or to triage existing findings (delimit_security_deliberate).

Sibling contrast: delimit_security_scan is the read-only baseline scanner; delimit_security_ingest accepts external tool output; delimit_security_deliberate triages findings via multi-model panel; this one runs the audit AND auto-chains evidence collection, governance task creation, and notification on criticals.

LED-1278: by default the scanner skips test directories (tests/, tests/, spec/, fixtures/, *_test.py, *.test.tsx, etc.) and suppresses well-known dummy values (AWS canonical example, alphabet-pattern GitHub tokens, leading-1234567890 Slack tokens, trivial JWTs, generic placeholder dict values). Pass include_tests=True to scan test trees too — useful for repos that ship real secrets in fixture files (rare, but legitimate).

Side effects: writes an evidence bundle (always, best-effort). On critical findings, creates a governance task via the governance engine and sends a webhook notification. Optional: SNYK_TOKEN or Trivy in the environment enable enhanced scanning.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
targetNoRepository or file path to audit. Default "." (cwd)..
include_testsNoWhen True, scan test directories (tests/, __tests__/, spec/, fixtures/, etc.). Default False — test trees are skipped to avoid the canonical fixture-credential FP class (LED-1278).

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv4.7.9
    • addedInput schema / properties / include_tests
      Added value: +{
+  "default": false,
+  "description": "When True, scan test directories (tests/, __tests__/, spec/, fixtures/, etc.). Default False — test trees are skipped to avoid the canonical fixture-credential FP class (LED-1278).",
+  "type": "boolean"
+}
    • changedInput schema / properties / target / description
      Previous value: -"Repository or file path to audit."New value: +"Repository or file path to audit. Default \".\" (cwd)."
  2. Changed1 schema field changedv4.5.5
    • addedInput schema / properties / target / description
      Added value: +"Repository or file path to audit."
  3. First observedv0.1.0

TDQS

A4.8/5.0
Behavior5/5

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

The description goes well beyond the sparse annotations (readOnlyHint=false, destructiveHint=false) by disclosing concrete side effects: always writes an evidence bundle (best-effort), and on critical findings creates a governance task and sends a webhook notification. It also reveals the LED-1278 default behavior (skipping test dirs, suppressing dummy values) and the optional SNYK_TOKEN/Trivy enhancement. No contradiction with annotations — readOnlyHint=false correctly matches the disclosed side effects.

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

Conciseness4/5

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

The description is longer than average but earns its length with clear section headers (When to use, When NOT to use, Sibling contrast, Side effects) and front-loading of the core purpose. Minor redundancy exists between the 'When NOT to use' list and the 'Sibling contrast' paragraph, but the structure keeps it scannable.

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

Completeness5/5

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

For a tool with side effects and a deploy-gate role, the description covers everything needed for correct invocation: purpose, routing, default behavior quirks, side-effect conditions, and optional environment enhancements. The output schema exists, so return-value documentation is already covered by structured data, and the two parameters are fully described.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3. The description adds value beyond the schema by explaining the LED-1278 rationale for include_tests and giving guidance on when passing include_tests=True is legitimate (repos shipping real secrets in fixtures). This contextualizes both parameters beyond their raw schema definitions.

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

Purpose5/5

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

The description opens with a specific verb+resource and the defining differentiator: 'Audit security and auto-chain evidence + governance on critical findings.' It enumerates the concrete checks performed (dependency scanning, secret detection, dangerous-pattern checks, .env-in-git) and explicitly contrasts with siblings, so an agent can identify it immediately.

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

Usage Guidelines5/5

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

Has dedicated 'When to use' and 'When NOT to use' sections that name exact alternatives (delimit_security_scan, delimit_security_ingest, delimit_security_deliberate) and the conditions that select each. The sibling-contrast paragraph reinforces the routing decision. This is the ideal level of usage guidance.

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

Deploy Server

Other Tools