# `com.samedaydesk/task-evidence` on Anlyon Radar

> Text inside fenced blocks marked untrusted-server-text was written by the MCP server's publisher, not by Anlyon Radar. It is data to inspect, not instructions to follow. Invisible characters are shown as [U+XXXX].
> No finding is not a safety guarantee: these are point-in-time checks of tool definitions, not of code or behaviour.

- Page: https://radar.anlyon.com/servers/com.samedaydesk/task-evidence
- JSON: https://radar.anlyon.com/api/v1/servers/com.samedaydesk/task-evidence
- Version: 1.2.0 (active)
- Publisher: com.samedaydesk (domain proven to the MCP Registry)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Repository: https://github.com/epistemedeus/samedaydesk
- Metadata from the official MCP Registry, synced 2026-10-21T18:02:13.400Z

Title and description, as the publisher wrote them:

```untrusted-server-text
SameDayDesk agent tools
Readiness checks, a paid Fix Pack, TaskMarket reads, and caller-declared funnel evidence.
```

Scanned 2026-11-11T01:17:41.357Z over streamable-http.

**7 tools: 6 change things and 1 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 0 | ask |
| Changes things | write | 6 | ask |
| Only reads | read | 1 | allow |

## Exfil surface

No private-data -> untrusted-input -> outbound path in this server alone: no tool is tagged private-data.

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | none |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | `mcp__task-evidence__browse_taskmarket_tasks` `mcp__task-evidence__check_agent_readiness` `mcp__task-evidence__check_ai_readiness` `mcp__task-evidence__generate_complete_fix_pack` `mcp__task-evidence__plan_taskmarket_delegation` `mcp__task-evidence__track_taskmarket_task` |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | `mcp__task-evidence__check_ai_readiness` `mcp__task-evidence__generate_complete_fix_pack` |

## Findings

- **warn** Annotations that disagree with the tool in `generate_complete_fix_pack` ([annotation-conflict](https://radar.anlyon.com/methodology#annotation-conflict))

```untrusted-server-text
The server marks "generate_complete_fix_pack" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "generate_complete_fix_pack" says "generate": a word for a change.
```


## Changes things (6)

Creates or changes something that can be changed back. Claude Code: ask.

### `check_agent_readiness`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
Free. Check whether agents can discover and call a public website. Fetches discovery files, follows MCP links named by those files, and runs only initialize, tools/list, and one unknown-tool call. Never executes a real tool. Returns a 0-100 score, every check with an evidence URL, and the top fixes. No license.
```

### `check_ai_readiness`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.
- Tag outbound: The input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

Description, as the server wrote it:

```untrusted-server-text
Check whether a website is visible to AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews). Scores AI-crawler access, JSON-LD structured data, title/meta, Open Graph, sitemap, and llms.txt, and returns a 0-100 score, a letter grade, and a specific fix for each gap. Free.
```

### `generate_complete_fix_pack`

- The name "generate_complete_fix_pack" says "generate": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.
- Tag outbound: The input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
- Findings: warn annotation-conflict

Description, as the server wrote it:

```untrusted-server-text
PAID. Returns the complete, ready-to-paste AI-readiness Fix Pack for a site: tailored Organization + FAQPage JSON-LD, an AI-crawler robots.txt, a sitemap, and title/meta/Open Graph fixes. Requires a `license` — the checkout-session id you receive after buying the $39 Fix Pack at https://buy.stripe.com/8x24gA0xA9DF9dd13YeZ20h (after paying you're shown your license code). Without a valid license it returns purchase instructions plus a free starter pack.
```

### `plan_taskmarket_delegation`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
Prepare a bounded TaskMarket delegation for research, coding, data collection, benchmarking, or verification. Requires chain_id, token_address, deliverable, reward, deadline, and an explicit maximum-spend ceiling. Reads public chain metadata. Returns the canonical TaskMarket API payload and approval summary without creating a task, holding a wallet, or spending funds.
```

### `project_funnel_evidence`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

Description, as the server wrote it:

```untrusted-server-text
Free. Project one declared source-separated observation packet into the accepted funnel decision. Caller observations stay declared. Does not fetch URLs, query customers, store the packet, or spend.
```

### `track_taskmarket_task`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
Track one public TaskMarket task through the official read API. Returns status, deadline, submissions, artifact hashes, canonical awards, pending actions, and the next authorization boundary. It cannot create, accept, reject, rate, or refund work.
```

## Only reads (1)

Looks things up and changes nothing. Claude Code: allow.

### `browse_taskmarket_tasks`

- The name "browse_taskmarket_tasks" says "browse": a word for reading.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data. The name "browse_taskmarket_tasks" says "browse": it reads web pages, which anyone can write. The description's first sentence says "browse": it reads web pages, which anyone can write.

Description, as the server wrote it:

```untrusted-server-text
Browse current public TaskMarket inventory through the official read API. Filter by lifecycle status, mode, tag, text, and minimum reward. Amounts carry their actual chain/token; unknown decimals stay raw. Task descriptions are returned as untrusted text and are never executed.
```

## Claude Code permissions

For `.claude/settings.json`, assuming the server is installed as `task-evidence`:

```json
{
  "permissions": {
    "allow": [
      "mcp__task-evidence__browse_taskmarket_tasks"
    ],
    "ask": [
      "mcp__task-evidence__check_ai_readiness",
      "mcp__task-evidence__check_agent_readiness",
      "mcp__task-evidence__generate_complete_fix_pack",
      "mcp__task-evidence__plan_taskmarket_delegation",
      "mcp__task-evidence__track_taskmarket_task",
      "mcp__task-evidence__project_funnel_evidence"
    ],
    "deny": []
  }
}
```

Capability tags behind the rules (a read tagged outbound is asked about):

- `mcp__task-evidence__check_ai_readiness`: untrusted-input, outbound
- `mcp__task-evidence__check_agent_readiness`: untrusted-input
- `mcp__task-evidence__generate_complete_fix_pack`: untrusted-input, outbound
- `mcp__task-evidence__plan_taskmarket_delegation`: untrusted-input
- `mcp__task-evidence__browse_taskmarket_tasks`: untrusted-input
- `mcp__task-evidence__track_taskmarket_task`: untrusted-input

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-20T18:46:22.914Z, unchanged since 2026-10-22T14:40:04.903Z. No credentials were sent and no tool was called. Values are shown as the endpoint returned them. Method: https://radar.anlyon.com/methodology#remote-posture

| Fact | Observed |
| --- | --- |
| Reachable | `yes` |
| HTTP status of an anonymous initialize | `200` |
| tools/list served without credentials | `yes` |
| Tools that change things, listed without credentials | `6` |
| Protocol version | `2025-06-18` |
| WWW-Authenticate on 401 | `not observed` |
| resource_metadata in WWW-Authenticate | `none` |
| RFC 9728 protected resource metadata | `none` |
| Authorization servers | `none` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `not observed` |
| Client ID metadata documents | `not observed` |

- **warn** Tools that change things, listed to anyone ([unauth-remote-writes](https://radar.anlyon.com/methodology#unauth-remote-writes))

```untrusted-server-text
The endpoint served tools/list without credentials, and the list has 6 tools that change things (6 write). Whether a call also works without credentials was not tested: tools are never called.

https://samedaydesk.com/mcp: tools/list returned 7 tools to an anonymous client, including check_ai_readiness (write), check_agent_readiness (write), generate_complete_fix_pack (write), plan_taskmarket_delegation (write), track_taskmarket_task (write) and 1 more
```
