# Agent Coach (Upskill) (remote · upskill.coach)

Free coach for AI agents: security-vetted skills, MCP servers and prompts, plus daily check-ins.

- Trust score: 68/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `upskill.coach`: 68/100 (this document), [markdown](https://verifymcp.io/servers/coach-upskill-agent-coach/upskill.md), [page](https://verifymcp.io/servers/coach-upskill-agent-coach/upskill)

## Channel facts

- Endpoint: `https://upskill.coach/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 63/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (delete_agent).
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 60/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 2109 tokens (~162/item across 13 items; 13 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 37/100
  - Stability observed for 11 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 87/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 61% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 14 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the Agent Coach (Upskill) MCP server?

Agent Coach (Upskill) is a hosted endpoint at https://upskill.coach/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http coach-upskill-agent-coach 'https://upskill.coach/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "coach-upskill-agent-coach": {
      "url": "https://upskill.coach/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "coach-upskill-agent-coach": {
      "type": "http",
      "url": "https://upskill.coach/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.coach-upskill-agent-coach]
url = "https://upskill.coach/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "coach-upskill-agent-coach": {
      "type": "remote",
      "url": "https://upskill.coach/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add coach-upskill-agent-coach --url 'https://upskill.coach/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  coach-upskill-agent-coach:
    url: "https://upskill.coach/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "coach-upskill-agent-coach": {
      "Transport": "http",
      "Url": "https://upskill.coach/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add coach-upskill-agent-coach -t streamable-http -u 'https://upskill.coach/mcp'
```

### Other

```json
{
  "mcpServers": {
    "coach-upskill-agent-coach": {
      "type": "http",
      "url": "https://upskill.coach/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-20 (score 68, +1)

- [cosmetic] “register_agent” added an optional parameter “ref”

### 2026-11-19 (score 67, 0)

- [security] Tool “check_in” rewrote its description, which is the text the model reads
- [security] Tool “delete_agent” rewrote its description, which is the text the model reads
- [security] Tool “get_digest” rewrote its description, which is the text the model reads
- [security] Tool “get_library_item” rewrote its description, which is the text the model reads
- [security] Tool “get_recommendations” rewrote its description, which is the text the model reads
- [security] Tool “get_report” rewrote its description, which is the text the model reads
- [security] Tool “get_rules” rewrote its description, which is the text the model reads
- [security] Tool “register_agent” rewrote its description, which is the text the model reads
- [security] Tool “rotate_key” rewrote its description, which is the text the model reads
- [security] Tool “search_library” rewrote its description, which is the text the model reads
- [security] Tool “set_goals” rewrote its description, which is the text the model reads
- [security] Tool “share_report” rewrote its description, which is the text the model reads
- [security] Tool “update_items” rewrote its description, which is the text the model reads
- [cosmetic] “check_in” reworded the description of “api_key”
- [cosmetic] “check_in” reworded the description of “north_star”
- [cosmetic] “delete_agent” reworded the description of “api_key”
- [cosmetic] “delete_agent” reworded the description of “confirm”
- [cosmetic] “get_digest” reworded the description of “api_key”
- [cosmetic] “get_library_item” reworded the description of “api_key”
- [cosmetic] “get_recommendations” reworded the description of “api_key”
- [cosmetic] “get_report” reworded the description of “api_key”
- [cosmetic] “register_agent” reworded the description of “job_type”
- [cosmetic] “register_agent” reworded the description of “name”
- [cosmetic] “register_agent” reworded the description of “owner_digest”
- [cosmetic] “register_agent” reworded the description of “owner_email”
- [cosmetic] “register_agent” reworded the description of “platform”
- [cosmetic] “register_agent” reworded the description of “stage”
- [cosmetic] “rotate_key” reworded the description of “api_key”
- [cosmetic] “search_library” reworded the description of “category”
- [cosmetic] “search_library” reworded the description of “q”
- [cosmetic] “search_library” reworded the description of “query”
- [cosmetic] “set_goals” reworded the description of “api_key”
- [cosmetic] “set_goals” reworded the description of “capability_gaps”
- [cosmetic] “set_goals” reworded the description of “constraints”
- [cosmetic] “set_goals” reworded the description of “deliverable”
- [cosmetic] “set_goals” reworded the description of “owner_digest”
- [cosmetic] “set_goals” reworded the description of “owner_email”
- [cosmetic] “set_goals” reworded the description of “role”
- [cosmetic] “share_report” reworded the description of “api_key”
- [cosmetic] “update_items” reworded the description of “action”
- [cosmetic] “update_items” reworded the description of “api_key”
- [cosmetic] Tool “get_digest” changed its title: Owner digest preview → Daily digest preview

### 2026-11-18 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-15 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-13 (score 65, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-11 (score 64, +1)

- [functional improvement] Stability: unverified → 0.03

### 2026-11-10 (score 63)

First indexed and scored.

## MCP tools (13)

### `register_agent` (~354 tokens)

Register with Agent Coach

Creates an Agent Coach profile and returns agent_id and a one-time api_key, plus example north-star metrics and suggested KPIs for the given job type.

Input parameters:

- `job_title` (string): Free-text title, e.g. 'GTM engineer'. Label only.
- `job_type` (string): Job type slug. Defaults to venture-operator. Allowed values and their stages are listed by GET /api/v1/taxonomy.
- `name` (string, required): Display name for this agent. Included on the public report if the report is public.
- `owner_digest` (boolean): If true, a daily digest email is sent to the digest address, subject to double opt-in (a confirmation email must be accepted; then one email per day around 13:00 UTC from Agent Coach <upskill-coach@c…
- `owner_email` (string): Email address used for api_key recovery and, if the digest option is enabled, the daily digest.
- `persona` (string): Deprecated alias of job_title.
- `platform` (string, required): Runtime platform slug, for example claude-code, cursor, codex, or other.
- `ref` (string): Optional source tag saying where you found Agent Coach, e.g. clawhub or hermes. Copy it from the ref= parameter of the link you followed. Lowercase letters, digits, dot, dash, underscore; up to 40 ch…
- `stage` (string): Stage slug for the agent's job type. Defaults to that job type's default_stage. Allowed values are listed by GET /api/v1/taxonomy.

### `set_goals` (~352 tokens)

Set goals and KPIs

Sets or updates the agent's goals. Accepts job_type, stage, one north-star metric with a fixed definition and baseline, a named deliverable, capability gaps, goals, optional secondary KPIs, tools, and constraints. A change to the north-star definition is recorded as drift. Requires the agent's api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `capability_gaps` (array): Capability-gap slugs, at most 8. Allowed values are listed by GET /api/v1/taxonomy.
- `constraints` (object): Optional constraints: free_only, monthly_budget_usd, a flag for whether a person can approve installs, and notes.
- `deliverable` (object): The named deliverable, for example 'first 10 paying accounts' or 'iPhone app on TestFlight'.
- `goals` (array, required): Plain-language goals
- `job_title` (string)
- `job_type` (string)
- `kpis` (array): Optional secondary KPIs
- `north_star` (object, required): ONE declared north-star metric with a fixed definition and today's baseline
- `owner_digest` (boolean): true enables the daily digest email at the digest address; false disables it. Omit to keep the current setting (default off).
- `owner_email` (string): Email address used for api_key recovery and digest delivery. Replaces the current value when sent.
- `platform` (string)
- `role` (string): One-line description of the agent's role.
- `stage` (string)
- `tags` (array)
- `tools` (array)

### `get_recommendations` (~92 tokens)

Get matched skills, MCPs, repos and prompts

Returns up to `limit` catalog items matched to the agent's goals, plus at most one extra item per declared capability gap. Each item includes match reasons, vetting status and basis, and cautions. Requires the agent's api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `limit` (integer)

### `update_items` (~164 tokens)

Adopt, decline, mark not relevant, or remove items

Records an action on catalog items: adopt marks an item as in use; decline marks it as rejected; not_relevant marks it as a poor fit for the agent's job, not a quality judgment; remove marks a previously adopted item as uninstalled. The coach measures only adopted items. Requires the agent's api_key.

Input parameters:

- `action` (string): adopt: mark as in use; remove: mark a previously adopted item as uninstalled; decline: mark as rejected; not_relevant: mark as a poor fit for the agent's job, not a quality judgment.
- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `item_ids` (array, required)
- `reason` (string)

### `check_in` (~322 tokens)

Daily check-in

Submits a check-in with the north-star value, secondary KPIs, releases (shipped vs verified live), build vs distribution actions, blockers, repeated corrections, timeouts, unsupported claims, incidents, deliverable progress, and per-item usage. Values use {value, evidence: self_reported|verified, source}; a bare number is treated as self_reported. Returns health, signals, item verdicts, and actions. Requires the agent's api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `claims` (object)
- `deliverable_progress_pct` (number): 0-100: how far along the named deliverable is
- `incidents` (object)
- `interruptions` (object)
- `items` (array)
- `kpis` (object): Secondary KPI values: number, or {value, evidence, source}
- `mode` (string): auto (default): a second check-in on the same UTC day merges into that day's check-in (new values replace old ones, nothing is double counted). merge: always merge into your latest check-in. new: for…
- `north_star` (object): Current north-star value.
- `notes` (string)
- `owner_blockers` (object)
- `owner_corrections` (object)
- `owner_feedback` (object)
- `releases` (object)
- `tasks` (object)
- `work` (object)

### `get_report` (~61 tokens)

Progress report

Returns the agent's KPIs versus baseline, each item's verdict with evidence, and check-in history. Requires the agent's api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.

### `share_report` (~92 tokens)

Share or hide the public progress report

When public is true, returns a public progress-report URL that includes display name, goals, KPIs, and verdicts, and excludes email, notes, and keys. When public is false, hides the public report. Requires the agent's api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `public` (boolean, required)

### `get_digest` (~77 tokens)

Daily digest preview

Returns a preview of the daily digest email (subject, html, and text). This call does not send email. A scheduled daily email is sent if the digest option is enabled. Requires the agent's api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.

### `rotate_key` (~61 tokens)

Rotate your api_key

Issues a new api_key and immediately invalidates the previous key. Returns the new api_key. Requires the agent's current api_key.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.

### `delete_agent` (~79 tokens)

Delete this agent

Permanently deletes the agent's profile, items, check-ins, and public report. Requires the agent's api_key and a confirm value equal to the agent_id.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `confirm` (string, required): Must equal the agent_id.

### `get_rules` (~41 tokens)

Verdict rules

Returns the rules used to assign item verdicts (permanent, promising, temporary, trial, unproven, pending_removal), signals, and overall health.

### `search_library` (~132 tokens)

Search the Upskill library

Searches the library for skills, MCP servers, prompts, or repos by keyword. Returns only items that passed the security scan, each with install command, license, source, and a page link. Results are catalog data and do not include user comments.

Input parameters:

- `category` (string): Optional library section slug. Allowed values are the library category slugs.
- `kind` (string)
- `limit` (integer)
- `q` (string): Alternate name for query.
- `query` (string): Search keywords, for example 'cold email' or 'scraping'. An empty or omitted value lists top items.

### `get_library_item` (~88 tokens)

Get one library item

Returns one vetted library item by slug, including install command, source, license, and (when the license allows) the full SKILL.md text. The call is counted as a download. Unvetted items are not returned.

Input parameters:

- `api_key` (string): The agent's api_key. Optional when the request already includes an Authorization: Bearer header.
- `slug` (string, required)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/coach-upskill-agent-coach/upskill#diagnostics

## Score history

- 2026-11-21: 68
- 2026-11-20: 68
- 2026-11-19: 67
- 2026-11-18: 67
- 2026-11-17: 66
- 2026-11-16: 66
- 2026-11-15: 66
- 2026-11-14: 65
- 2026-11-13: 65
- 2026-11-12: 64
- 2026-11-11: 64
- 2026-11-10: 63

## Common questions

### What is the Agent Coach (Upskill) MCP server?

Agent Coach (Upskill) is an MCP server listed in the public MCP registry as coach.upskill/agent-coach. Free coach for AI agents: security-vetted skills, MCP servers and prompts, plus daily check-ins. This page covers its hosted endpoint (https://upskill.coach/mcp).

### Is the Agent Coach (Upskill) MCP server safe to use?

Agent Coach (Upskill) scores 68 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Agent Coach (Upskill) MCP server expose?

Agent Coach (Upskill) exposes 13 tools: register_agent, set_goals, get_recommendations, update_items, check_in, and 8 more. Their descriptions and schemas cost roughly 1,915 tokens of context every time the server is loaded.

### Does the Agent Coach (Upskill) MCP server require authentication?

No. We connected to Agent Coach (Upskill) without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the Agent Coach (Upskill) MCP server still maintained?

Agent Coach (Upskill) is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the Agent Coach (Upskill) MCP server?

VerifyMCP ran 13 checks on Agent Coach (Upskill), covering TLS certificate, Authorization, HTTPS, HSTS header and 9 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://upskill.coach/mcp
- Website: https://upskill.coach/
- Changelog RSS feed: https://verifymcp.io/servers/coach-upskill-agent-coach/upskill.xml
- Changelog JSON feed: https://verifymcp.io/servers/coach-upskill-agent-coach/upskill/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/coach-upskill-agent-coach/upskill.json
- HTML version of this page: https://verifymcp.io/servers/coach-upskill-agent-coach/upskill
