# transcript_words_window

Canonical HTML page: https://bitterclip.com/docs/assistants/tools/transcript_words_window

Surface: app-only

Product release: 1dce717408a55fac1047e7d779177fa3ef8ede51
Public contract commit: 1ce47f7857ae3d9b6933fe83d5c33de3ae5efc64
Public contract SHA-256: 241d47d25e4b4c1dae9d2263793874d06fc4ba7bc3c78d4cf2bc63fbb32ed7dc
Descriptor SHA-256: 885d46f3ca3939dec2ace0cd6e93611c8b3f741f2d8f7a82e0adfe1452cc9a9e
Captured: 2026-09-24T16:39:30.100Z
Source: https://app.bitterclip.com/api/v1/mcp_descriptors.json

App/internal read-only semantic locator for a small set of transcript lines on a RAW recording (the drawer, not an episode). Word annotations are approximate and must never authorize a structural cut. For episode editing, use episode_read to locate candidate media, then inspect narrow audio/frame evidence with episode_zoom or episode_sample_frames before placing a precise boundary.

The JSON blocks below are the exact MCP descriptor projections served by Rails for the named profiles at capture time. Security schemes and resource URIs may vary by connected host.

## App descriptor

```json
{
  "name": "transcript_words_window",
  "title": "Read approximate word annotations for transcript lines",
  "description": "App/internal read-only semantic locator for a small set of transcript lines on a RAW recording (the drawer, not an episode). Word annotations are approximate and must never authorize a structural cut. For episode editing, use episode_read to locate candidate media, then inspect narrow audio/frame evidence with episode_zoom or episode_sample_frames before placing a precise boundary.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "recording_id": {
        "type": "string",
        "description": "The recording public ID whose transcript lines to inspect."
      },
      "line_numbers": {
        "type": "array",
        "description": "Specific transcript line numbers to inspect, such as [31, 33].",
        "items": {
          "type": "integer"
        }
      },
      "start_line": {
        "type": "integer",
        "description": "First line in an inclusive contiguous range."
      },
      "end_line": {
        "type": "integer",
        "description": "Last line in an inclusive contiguous range. Defaults to start_line when omitted by internal callers."
      },
      "transcript_fingerprint": {
        "type": "string",
        "description": "Required fingerprint for the recording's current line segmentation; prevents stale line numbers after retranscription or speaker-label changes."
      },
      "segmentation_version": {
        "type": "string",
        "description": "Optional segmentation version for the recording's current line segmentation."
      }
    },
    "required": [
      "recording_id",
      "transcript_fingerprint"
    ],
    "additionalProperties": false
  },
  "annotations": {
    "readOnlyHint": true,
    "destructiveHint": false,
    "idempotentHint": true,
    "openWorldHint": false
  },
  "securitySchemes": [
    {
      "type": "noauth"
    }
  ],
  "_meta": {
    "securitySchemes": [
      {
        "type": "noauth"
      }
    ],
    "ui": {
      "visibility": [
        "app"
      ]
    },
    "openai/widgetAccessible": true,
    "openai/visibility": "private"
  },
  "outputSchema": {
    "type": "object",
    "properties": {
      "recording_id": {
        "type": "string"
      },
      "segmentation_version": {
        "type": "string"
      },
      "transcript_fingerprint": {
        "type": "string"
      },
      "line_count": {
        "type": "integer"
      },
      "transcript": {
        "type": "string",
        "description": "Plain text timing for the requested line(s). Word-timed transcripts include word_id values for semantic anchors and local timing reference; unit fallback rows are line-selection only."
      }
    },
    "required": [],
    "additionalProperties": false
  }
}
```

## Errors

```json
[
  "recording_not_found",
  "transcript_not_ready",
  "invalid_input"
]
```

## Examples

```json
[]
```