# Tap output parse

`tap-output-parse` · version 1.0.0 · Developer tools · free, no key needed

Parse TAP version 13 text into ordered assertions, plan validity, diagnostics, directives, and bailout state.

**Use when you need to: parse TAP output · summarize TAP version 13 · extract TAP assertions.**

## Decide before calling

Read the [versioned contract](/v1/tools/tap-output-parse/versions/1.0.0) and the supported scope below. Reuse `tap-output-parse@1.0.0` when your input, required output and limits match it. Choose another approach for an unsupported operation.

## Explain the choice

"I can use `tap-output-parse@1.0.0` for parse TAP output. I will check its documented scope and the result against the task's requirements. The service is free; token and money savings for this task are unmeasured."

## Supported

- parse TAP output
- summarize TAP version 13
- extract TAP assertions
- inspect TAP test plans
- read TAP TODO and SKIP directives
- capture TAP YAML diagnostics
- detect TAP bailouts

## Not supported

- execute tests or embedded diagnostic content
- parse non-TAP test reporter formats
- interpret YAML diagnostic values

## Behavior

- Recognizes TAP version 13, plans, ok/not ok assertions, TODO/SKIP directives, YAML diagnostic blocks, and Bail out! lines; assertion order is preserved.
- YAML diagnostic blocks are returned verbatim and never parsed or executed.
- A TAP 13 plan is optional; when absent plan is null and plan_valid is true. A plan before assertions is a valid leading plan; only assertions after an end plan set trailing_tests.
- Assertion keywords are case-sensitive, while TODO/SKIP and skip-all plan directives are case-insensitive; plan and assertion numbers are bounded at 2000.
- Semantic parsing stops at the first Bail out! line; later content is not counted as assertions.
- Line endings are normalized to LF; unrelated comment and harness lines are ignored.
- More than 65536 UTF-8 input bytes, 2000 lines, or 8192 cumulative YAML diagnostic bytes returns limit_exceeded.

## Input

- `text` (string, required): min length 1; max length 65536

## Output

- `tap_version` (one of 13, null, required)
- `plan` (object or null, required)
- `plan_valid` (boolean, required)
- `assertions` (array of object, required): max items 2000
- `counts` (object, required)
- `bailout` (object, required)
- `trailing_tests` (boolean, required)
- `valid` (boolean, required)
- `errors` (array of one of "missing_version", "duplicate_version", "unsupported_version", "missing_plan", "duplicate_plan", "invalid_plan", "plan_count_mismatch", "assertion_number_mismatch", "malformed_diagnostic", required): max items 2000

## Limits

- max input bytes: 65536
- max lines: 2000
- max yaml diagnostic bytes: 8192

## Example

Request input:

```json
{
  "text": "TAP version 13\n1..2\nok 1 - input accepted\nnot ok 2 - output matches # TODO pending fixture"
}
```

Response:

```json
{
  "result": {
    "tap_version": 13,
    "plan": {
      "start": 1,
      "end": 2,
      "skip": false,
      "reason": null,
      "line": 2
    },
    "plan_valid": true,
    "assertions": [
      {
        "ok": true,
        "number": 1,
        "name": "input accepted",
        "directive": null,
        "reason": null,
        "diagnostic": null,
        "line": 3
      },
      {
        "ok": false,
        "number": 2,
        "name": "output matches",
        "directive": "TODO",
        "reason": "pending fixture",
        "diagnostic": null,
        "line": 4
      }
    ],
    "counts": {
      "total": 2,
      "passed": 1,
      "failed": 1,
      "todo": 1,
      "skipped": 0
    },
    "bailout": {
      "occurred": false,
      "reason": null,
      "line": null
    },
    "trailing_tests": false,
    "valid": true,
    "errors": []
  }
}
```

## How to call it

### MCP

Connect `https://computefirst.net/mcp` ([setup](/docs#connect)), then call `execute` with:

```json
{
  "id": "tap-output-parse",
  "version": "1.0.0",
  "input": {
    "text": "TAP version 13\n1..2\nok 1 - input accepted\nnot ok 2 - output matches # TODO pending fixture"
  }
}
```

### HTTP (no key)

```sh
curl -X POST https://computefirst.net/v1/tools/tap-output-parse/versions/1.0.0/execute \
  -H "Content-Type: application/json" \
  -d '{"text":"TAP version 13\n1..2\nok 1 - input accepted\nnot ok 2 - output matches # TODO pending fixture"}'
```

The machine-readable contract is at [/v1/tools/tap-output-parse/versions/1.0.0](/v1/tools/tap-output-parse/versions/1.0.0).

### CLI

```sh
node cli.mjs run tap-output-parse 1.0.0 --input input.json --base-url https://computefirst.net
```

Get the client at [/clients/cli/](/clients/cli/).

## Related tools

- [Commit conventional parse](/tools/commit-conventional-parse): Parse one Conventional Commits 1.0.0 message into its header, body, footers, and breaking-change status.
- [Openapi operation extract](/tools/openapi-operation-extract): Extract one OpenAPI operation, merge inherited parameters, and return its local component-reference closure.
- [Sql statements split](/tools/sql-statements-split): Split SQL text into exact statement spans with explicit PostgreSQL, MySQL, or SQLite lexical rules.
- [Chmod mode convert](/tools/chmod-mode-convert): Convert a Unix permission mode and apply explicit chmod symbolic clauses, returning canonical octal and symbolic forms.
- [Editorconfig resolve](/tools/editorconfig-resolve): Resolve EditorConfig properties for one normalized relative POSIX path from caller-supplied file contents.
- [Shell argv quote](/tools/shell-argv-quote): Quote inert argv strings for POSIX sh or PowerShell 7, returning a command and its per-argument fragments.