Skip to content

CompletionKit MCP server

com.completionkit/evalscompletionkit.comstreamable-httpv1.0.0Scanned

Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.

54 tools: 8 can delete or cancel things, 2 do something that can't be undone, 26 change things and 18 only read.

Tools: 8 delete, 2 irreversible, 26 write, 18 read.
  • Reads private data · 2
  • Reads text others write · 0
  • Can send data out · 3
  • 5 findings

Tools listed without sign-in. Remote posture

With Anlyon

8 tools here can delete or cancel things and 2 do something that can't be undone. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the CompletionKit MCP server can do

The CompletionKit MCP server is published by completionkit.com, a domain the MCP Registry verified. It has 54 tools. It can delete or cancel things (8 tools: prompts_delete, runs_delete, datasets_delete, metrics_delete and 4 more), do things that cannot be undone, such as sending, paying or publishing (prompts_publish and metric_versions_publish) and create or change data (26 tools: prompts_create, prompts_update, prompts_suggest_improvement, runs_list and 22 more). 18 tools only read.

2 tools read private data (provider_credentials_list and provider_credentials_get); 3 tools can send data out of your system (prompts_publish, datasets_create_from_url and metric_versions_publish).

The fixed checks matched 5 findings (credentials-mention). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 10, 2026, version 1.0.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

54 tools

Deletes or cancels 8Claude Code: deny
  • datasets_deleteDelete a datasetNo findings

    Why it is in this group

    • The name "datasets_delete" says "delete": a word for removing or ending something.
    Description, as the server wrote it
    Delete a dataset
  • metric_groups_deleteDelete a metric groupNo findings

    Why it is in this group

    • The name "metric_groups_delete" says "delete": a word for removing or ending something.
    Description, as the server wrote it
    Delete a metric group
  • metric_versions_dismissDestroy a draft MetricVersion (use for either source: 'edit' or source: 'suggestion').No findings

    Why it is in this group

    • The name "metric_versions_dismiss" says "dismiss": a word for a change.
    • The description's first verb is "destroy": a word for removing or ending something.
    Description, as the server wrote it
    Destroy a draft MetricVersion (use for either source: 'edit' or source: 'suggestion'). Published versions are refused. To demote a published version, publish a different one as current instead.
  • metrics_deleteDelete a metricNo findings

    Why it is in this group

    • The name "metrics_delete" says "delete": a word for removing or ending something.
    Description, as the server wrote it
    Delete a metric
  • prompts_deleteDelete a promptNo findings

    Why it is in this group

    • The name "prompts_delete" says "delete": a word for removing or ending something.
    Description, as the server wrote it
    Delete a prompt
  • provider_credentials_deleteDelete a provider credential1 finding

    Why it is in this group

    • The name "provider_credentials_delete" says "delete": a word for removing or ending something.

    Findings

    Description, as the server wrote it
    Delete a provider credential
  • runs_deleteDelete a runNo findings

    Why it is in this group

    • The name "runs_delete" says "run": a word for a change.
    • The description's first verb is "delete": a word for removing or ending something.
    Description, as the server wrote it
    Delete a run
  • tags_deleteDelete a tag.No findings

    Why it is in this group

    • The name "tags_delete" says "delete": a word for removing or ending something.
    Description, as the server wrote it
    Delete a tag. Removes the tag from every linked metric, prompt, run, and dataset.
Can't be undone 2Claude Code: ask
  • metric_versions_publishPublish a MetricVersion as the live version of its metric.Can send data outNo findings

    Why it is in this group

    • The name "metric_versions_publish" says "publish": a word for a change that another call does not undo.

    What it can reach

    • outboundThe name "metric_versions_publish" says "publish": it sends or publishes something to other people.
    Description, as the server wrote it
    Publish a MetricVersion as the live version of its metric. Works for both 'draft → published' and 'revert to an older published version → current'. Transactionally flips current, demotes peers, and writes the version's instruction + rubric_bands back onto the metric so the judge grades against it.
  • prompts_publishPublish a prompt version, making it the current versionCan send data outNo findings

    Why it is in this group

    • The name "prompts_publish" says "publish": a word for a change that another call does not undo.

    What it can reach

    • outboundThe name "prompts_publish" says "publish": it sends or publishes something to other people.
    Description, as the server wrote it
    Publish a prompt version, making it the current version
Changes things 26Claude Code: ask
  • agreements_createUpsert an agreement for (run, response, metric, created_by).No findings

    Why it is in this group

    • The name "agreements_create" says "create": a word for a change.
    Description, as the server wrote it
    Upsert an agreement for (run, response, metric, created_by). Verdict is one of agree, disagree, borderline. corrected_score (1..5) is required when verdict is 'disagree'.
  • datasets_createCreate a dataset with CSV data.No findings

    Why it is in this group

    • The name "datasets_create" says "create": a word for a change.
    Description, as the server wrote it
    Create a dataset with CSV data. First row is the header. Two column names are recognized specially: "expected_output" is each row's answer key (ground truth) given to the judge and to checks that compare against the row's expected value, and "actual_output" is a pre-made output to score in a prompt-less run. Both are overridable per run (expected_column / output_column). Every column is also available to the prompt as a variable.
  • datasets_create_from_urlCreate a dataset by downloading CSV from a URL instead of inlining it.Can send data outNo findings

    Why it is in this group

    • The name "datasets_create_from_url" says "create": a word for a change.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Create a dataset by downloading CSV from a URL instead of inlining it. Use this for large datasets: pass a public http(s) URL and the server fetches the CSV directly, so the data never has to pass through the tool-call arguments. The URL is SSRF-checked and the download is capped at 10MB. First row is the header; the "expected_output" (answer key) and "actual_output" (pre-made output) columns are recognized specially, overridable per run.
  • datasets_updateUpdate a datasetNo findings

    Why it is in this group

    • The name "datasets_update" says "update": a word for a change.
    Description, as the server wrote it
    Update a dataset
  • judges_replayCreate a scoring run for the current judge over a dataset's existing outputs (wraps runs_create with prompt_id omitted and output_column supplied).No findings

    Why it is in this group

    • The description's first verb is "create": a word for a change.
    Description, as the server wrote it
    Create a scoring run for the current judge over a dataset's existing outputs (wraps runs_create with prompt_id omitted and output_column supplied). This only sets up the run; call runs_generate to actually re-judge the outputs so you can compare against human verdicts.
  • metric_groups_createCreate a metric groupNo findings

    Why it is in this group

    • The name "metric_groups_create" says "create": a word for a change.
    Description, as the server wrote it
    Create a metric group
  • metric_groups_updateUpdate a metric groupNo findings

    Why it is in this group

    • The name "metric_groups_update" says "update": a word for a change.
    Description, as the server wrote it
    Update a metric group
  • metrics_createCreate a metric with evaluation criteria.No findings

    Why it is in this group

    • The name "metrics_create" says "create": a word for a change.
    Description, as the server wrote it
    Create a metric with evaluation criteria. For a deterministic check set metric_type:"check" and check_config. Per-kind required keys: value (contains/not_contains/equals/list_overlap/numeric_equals), pattern (regex), json_path+expected (json_path_equals), min and/or max (length_bounds/numeric_bounds); valid_json takes no extra keys. target_path is required when target is json_path; a target_path sent with no target sets target to json_path, and one sent with any other target is refused. target_path walks array indices, as in items.0.name; any path may start at the root with $., as in $.items.0.name. list_overlap grades a list-valued field and stores a fraction rather than a verdict: score_by is recall, precision, f1 or jaccard (recall by default), and min is the fraction that counts as a pass (1, a perfect score, by default). numeric_bounds and numeric_equals read a number, ignoring the thousands commas and currency symbol a figure is written with; a comma counts only as a thousands separator, with three digits after it and no decimal point before it, so 1,5 is not a number; numeric_equals takes a tolerance, either a plain number for a fixed amount or a percentage string such as 2% for a share of the expected value. A check with no case_sensitive matches case-insensitively, regex and list_overlap included; send case_sensitive: true to match case exactly. Before 0.29.0 a regex check read a missing case_sensitive as case-sensitive. Set compare_to:"expected" on contains, not_contains, equals, json_path_equals, list_overlap, numeric_equals to grade against each row's own expected_output (ground truth) instead of a constant; drop value (or expected, for json_path_equals) and add expected_path to dig into the answer key when it is JSON.
  • metrics_suggest_variantsAsk the model to rewrite the metric's judge instruction in N variants targeted at the recent disagreements.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Ask the model to rewrite the metric's judge instruction in N variants targeted at the recent disagreements. Each variant is saved as a draft MetricVersion with source="suggestion". Returns the persisted drafts. Stripe-metering hooks fire via ActiveSupport::Notifications under completion_kit.judge_suggestion.generated.
  • metrics_updateUpdate a metric.No findings

    Why it is in this group

    • The name "metrics_update" says "update": a word for a change.
    Description, as the server wrote it
    Update a metric. For a deterministic check set metric_type:"check" and check_config. Per-kind required keys: value (contains/not_contains/equals/list_overlap/numeric_equals), pattern (regex), json_path+expected (json_path_equals), min and/or max (length_bounds/numeric_bounds); valid_json takes no extra keys. target_path is required when target is json_path; a target_path sent with no target sets target to json_path, and one sent with any other target is refused. target_path walks array indices, as in items.0.name; any path may start at the root with $., as in $.items.0.name. list_overlap grades a list-valued field and stores a fraction rather than a verdict: score_by is recall, precision, f1 or jaccard (recall by default), and min is the fraction that counts as a pass (1, a perfect score, by default). numeric_bounds and numeric_equals read a number, ignoring the thousands commas and currency symbol a figure is written with; a comma counts only as a thousands separator, with three digits after it and no decimal point before it, so 1,5 is not a number; numeric_equals takes a tolerance, either a plain number for a fixed amount or a percentage string such as 2% for a share of the expected value. A check with no case_sensitive matches case-insensitively, regex and list_overlap included; send case_sensitive: true to match case exactly. Before 0.29.0 a regex check read a missing case_sensitive as case-sensitive. Set compare_to:"expected" on contains, not_contains, equals, json_path_equals, list_overlap, numeric_equals to grade against each row's own expected_output (ground truth) instead of a constant; drop value (or expected, for json_path_equals) and add expected_path to dig into the answer key when it is JSON.
  • promptfoo_importImport a promptfooconfig.yaml.No findings

    Why it is in this group

    • The name "promptfoo_import" says "import": a word for a change.
    Description, as the server wrote it
    Import a promptfooconfig.yaml. Creates a prompt, a dataset from the test vars, and metrics from the assert blocks (llm-rubric/g-eval become judge metrics; contains/equals/regex/is-json become deterministic check metrics). Returns a summary of what mapped and what was skipped and why; nothing is dropped silently.
  • prompts_createCreate a promptNo findings

    Why it is in this group

    • The name "prompts_create" says "create": a word for a change.
    Description, as the server wrote it
    Create a prompt
  • prompts_suggest_improvementSuggest an improved version of a prompt, grounded in a run's test results and judge feedback.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Suggest an improved version of a prompt, grounded in a run's test results and judge feedback. Analyzes the run's responses, scores, and reviews, then returns reasoning plus a rewritten template (preserving {{variables}}) and persists it as a Suggestion. Requires a run that has a prompt (not a scoring-only run).
  • prompts_updateUpdate a prompt.No findings

    Why it is in this group

    • The name "prompts_update" says "update": a word for a change.
    Description, as the server wrote it
    Update a prompt. If the prompt already has runs, this creates a new DRAFT version (current=false) rather than editing in place or publishing. Promote it with prompts_publish, so an agent's edits don't go live without a gate. If it has no runs, it is updated in place.
  • provider_credentials_createCreate a provider credential1 finding

    Why it is in this group

    • The name "provider_credentials_create" says "create": a word for a change.

    Findings

    Description, as the server wrote it
    Create a provider credential
  • provider_credentials_updateUpdate a provider credential.1 finding

    Why it is in this group

    • The name "provider_credentials_update" says "update": a word for a change.

    Findings

    • Info

      The description mentions "credential".

      Matched text

      description: "Update a provider credential. Changing provider or api_endpoint on a..."

      What this check looks for
    Description, as the server wrote it
    Update a provider credential. Changing provider or api_endpoint on a credential with a stored key requires api_key in the same call, so a stored key never moves to a new provider or endpoint.
  • runs_createCreate a run.No findings

    Why it is in this group

    • The name "runs_create" says "run": a word for a change.
    Description, as the server wrote it
    Create a run. Omit prompt_id and provide output_column to score existing outputs by grading a pre-existing dataset column instead of generating new ones.
  • runs_generateStart a run.No findings

    Why it is in this group

    • The name "runs_generate" says "run": a word for a change.
    Description, as the server wrote it
    Start a run. Required for every run, including score-only runs (no prompt): generates responses with the prompt when there is one, otherwise copies the graded dataset column and grades it.
  • runs_getGet a run by ID, including "metric_averages": a per-metric breakdown with each metric's average score (or pass rate for checks), how many rows it graded, and how many scored low.No findings

    Why it is in this group

    • The name "runs_get" says "run": a word for a change.
    Description, as the server wrote it
    Get a run by ID, including "metric_averages": a per-metric breakdown with each metric's average score (or pass rate for checks), how many rows it graded, and how many scored low. Use this to find the metric dragging a prompt down without listing responses.
  • runs_listList all runsNo findings

    Why it is in this group

    • The name "runs_list" says "run": a word for a change.
    Description, as the server wrote it
    List all runs
  • runs_regradeRe-grade a run's existing responses with its currently attached metrics, without regenerating.No findings

    Why it is in this group

    • The name "runs_regrade" says "run": a word for a change.
    Description, as the server wrote it
    Re-grade a run's existing responses with its currently attached metrics, without regenerating. Use after attaching or editing metrics on an already-generated run.
  • runs_rerunCreate and start a fresh copy of a run with the same prompt, dataset, metrics, and settings.No findings

    Why it is in this group

    • The name "runs_rerun" says "run": a word for a change.
    Description, as the server wrote it
    Create and start a fresh copy of a run with the same prompt, dataset, metrics, and settings. Use when the judge changed and you want a clean run instead of mixing versions.
  • runs_retry_failuresRe-run only the failed responses of a run, optionally limited to specific response ids via "only".No findings

    Why it is in this group

    • The name "runs_retry_failures" says "run": a word for a change.
    Description, as the server wrote it
    Re-run only the failed responses of a run, optionally limited to specific response ids via "only".
  • runs_updateUpdate a runNo findings

    Why it is in this group

    • The name "runs_update" says "run": a word for a change.
    Description, as the server wrote it
    Update a run
  • tags_createCreate a tag.No findings

    Why it is in this group

    • The name "tags_create" says "create": a word for a change.
    Description, as the server wrote it
    Create a tag. Color is auto-assigned.
  • tags_updateRename a tag.No findings

    Why it is in this group

    • The name "tags_update" says "update": a word for a change.
    Description, as the server wrote it
    Rename a tag.
Only reads 18Show 18 read-only tools
  • agreements_listList agreements.No findings

    Why it is in this group

    • The name "agreements_list" says "list": a word for reading.
    Description, as the server wrote it
    List agreements. Filter by run_id, response_id, metric_id, or created_by.
  • datasets_getGet a dataset by IDNo findings

    Why it is in this group

    • The name "datasets_get" says "get": a word for reading.
    Description, as the server wrote it
    Get a dataset by ID
  • datasets_listList all datasetsNo findings

    Why it is in this group

    • The name "datasets_list" says "list": a word for reading.
    Description, as the server wrote it
    List all datasets
  • judges_compareCompare two versions of one metric's agreement stats side by side.No findings

    Why it is in this group

    • The name "judges_compare" says "compare": a word for reading.
    Description, as the server wrote it
    Compare two versions of one metric's agreement stats side by side. Requires metric_id, metric_version_a_id, and metric_version_b_id (both versions must belong to that metric). Unavailable for check metrics.
  • metric_groups_getGet a metric group by IDNo findings

    Why it is in this group

    • The name "metric_groups_get" says "get": a word for reading.
    Description, as the server wrote it
    Get a metric group by ID
  • metric_groups_listList all metric groupsNo findings

    Why it is in this group

    • The name "metric_groups_list" says "list": a word for reading.
    Description, as the server wrote it
    List all metric groups
  • metric_versions_listList every MetricVersion (drafts + published) for a metric, newest first.No findings

    Why it is in this group

    • The name "metric_versions_list" says "list": a word for reading.
    Description, as the server wrote it
    List every MetricVersion (drafts + published) for a metric, newest first. Each row carries version_number, state, source, current flag, and timestamps.
  • metrics_getGet a metric by IDNo findings

    Why it is in this group

    • The name "metrics_get" says "get": a word for reading.
    Description, as the server wrote it
    Get a metric by ID
  • metrics_listList all metricsNo findings

    Why it is in this group

    • The name "metrics_list" says "list": a word for reading.
    Description, as the server wrote it
    List all metrics
  • prompts_getGet a prompt by IDNo findings

    Why it is in this group

    • The name "prompts_get" says "get": a word for reading.
    Description, as the server wrote it
    Get a prompt by ID
  • prompts_listList all promptsNo findings

    Why it is in this group

    • The name "prompts_list" says "list": a word for reading.
    Description, as the server wrote it
    List all prompts
  • provider_credentials_getGet a provider credential by ID (API key is not exposed)Reads private data1 finding

    Why it is in this group

    • The name "provider_credentials_get" says "get": a word for reading.

    What it can reach

    • private-dataThe name "provider_credentials_get" says "get" and "credential": it reads secrets.
    • private-dataThe description's first sentence says "get" and "credential": it reads secrets.

    Findings

    • Info

      The description mentions "credential".

      Matched text

      description: "Get a provider credential by ID (API key is not exposed)"

      What this check looks for
    Description, as the server wrote it
    Get a provider credential by ID (API key is not exposed)
  • provider_credentials_listList all provider credentials (API keys are not exposed)Reads private data1 finding

    Why it is in this group

    • The name "provider_credentials_list" says "list": a word for reading.

    What it can reach

    • private-dataThe name "provider_credentials_list" says "list" and "credential": it reads secrets.
    • private-dataThe description's first sentence says "list" and "credential": it reads secrets.

    Findings

    • Info

      The description mentions "credentials".

      Matched text

      description: "List all provider credentials (API keys are not exposed)"

      What this check looks for
    Description, as the server wrote it
    List all provider credentials (API keys are not exposed)
  • responses_getGet a specific responseNo findings

    Why it is in this group

    • The name "responses_get" says "get": a word for reading.
    Description, as the server wrote it
    Get a specific response
  • responses_listList responses for a run, in row order.No findings

    Why it is in this group

    • The name "responses_list" says "list": a word for reading.
    Description, as the server wrote it
    List responses for a run, in row order. Returns {total, limit, offset, returned, responses}. Defaults to 50 rows because full payloads are large: use "fields" to drop the bodies, "min_score"/"max_score" to isolate low scorers, and sort "score_asc" to read the worst rows first. For per-metric averages of the whole run use runs_get instead of aggregating here.
  • tags_getGet a tag by IDNo findings

    Why it is in this group

    • The name "tags_get" says "get": a word for reading.
    Description, as the server wrote it
    Get a tag by ID
  • tags_listList all tagsNo findings

    Why it is in this group

    • The name "tags_list" says "list": a word for reading.
    Description, as the server wrote it
    List all tags
  • usage_getGet this organization's plan usage and limits for the current billing period: runs and prompt fetches used, their limits, how many remain, and when the period resets.No findings

    Why it is in this group

    • The name "usage_get" says "get": a word for reading.
    Description, as the server wrote it
    Get this organization's plan usage and limits for the current billing period: runs and prompt fetches used, their limits, how many remain, and when the period resets. Call this to pre-check quota before starting runs. Runs are hard-blocked once the run limit is reached (with a small grace band), so a run over the limit will fail with run_limit_reached.

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

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

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out3
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:36 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
36
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 36 tools that change things (8 destructive, 2 irreversible, 26 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://completionkit.com/mcp: tools/list returned 54 tools to an anonymous client, including prompts_delete (destructive), runs_delete (destructive), datasets_delete (destructive), metrics_delete (destructive), metric_groups_delete (destructive) and 31 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

5 info

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.0.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http evals https://completionkit.com/mcp --header 'Authorization: <your-authorization>'
claude mcp add --transport http evals https://completionkit.com/mcp --header 'Authorization: <your-authorization>'

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__evals__prompts_list",
      "mcp__evals__prompts_get",
      "mcp__evals__responses_list",
      "mcp__evals__responses_get",
      "mcp__evals__datasets_list",
      "mcp__evals__datasets_get",
      "mcp__evals__metrics_list",
      "mcp__evals__metrics_get",
      "mcp__evals__metric_groups_list",
      "mcp__evals__metric_groups_get",
      "mcp__evals__metric_versions_list",
      "mcp__evals__provider_credentials_list",
      "mcp__evals__provider_credentials_get",
      "mcp__evals__tags_list",
      "mcp__evals__tags_get",
      "mcp__evals__agreements_list",
      "mcp__evals__judges_compare",
      "mcp__evals__usage_get"
    ],
    "ask": [
      "mcp__evals__prompts_create",
      "mcp__evals__prompts_update",
      "mcp__evals__prompts_publish",
      "mcp__evals__prompts_suggest_improvement",
      "mcp__evals__runs_list",
      "mcp__evals__runs_get",
      "mcp__evals__runs_create",
      "mcp__evals__runs_update",
      "mcp__evals__runs_generate",
      "mcp__evals__runs_regrade",
      "mcp__evals__runs_rerun",
      "mcp__evals__runs_retry_failures",
      "mcp__evals__datasets_create",
      "mcp__evals__datasets_update",
      "mcp__evals__datasets_create_from_url",
      "mcp__evals__metrics_create",
      "mcp__evals__metrics_update",
      "mcp__evals__metrics_suggest_variants",
      "mcp__evals__metric_groups_create",
      "mcp__evals__metric_groups_update",
      "mcp__evals__metric_versions_publish",
      "mcp__evals__provider_credentials_create",
      "mcp__evals__provider_credentials_update",
      "mcp__evals__tags_create",
      "mcp__evals__tags_update",
      "mcp__evals__agreements_create",
      "mcp__evals__judges_replay",
      "mcp__evals__promptfoo_import"
    ],
    "deny": [
      "mcp__evals__prompts_delete",
      "mcp__evals__runs_delete",
      "mcp__evals__datasets_delete",
      "mcp__evals__metrics_delete",
      "mcp__evals__metric_groups_delete",
      "mcp__evals__metric_versions_dismiss",
      "mcp__evals__provider_credentials_delete",
      "mcp__evals__tags_delete"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as evals. Review it before you use it.

Capability tags behind 5 rules
  • mcp__evals__prompts_publish outbound
  • mcp__evals__datasets_create_from_url outbound
  • mcp__evals__metric_versions_publish outbound
  • mcp__evals__provider_credentials_list private-data
  • mcp__evals__provider_credentials_get private-data

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.completionkit/evals/verify?hash=37766d4e223766f6ad95020eff5459075ed9b0904f581154102e54f6f42ae532" | jq -e '.changed == false'

Pinned hash 37766d4e2237, from Nov 10, 2026, 13:17 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.completionkit/evals/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.completionkit/evals
[![Anlyon Radar](https://radar.anlyon.com/badge/com.completionkit/evals.svg)](https://radar.anlyon.com/servers/com.completionkit/evals)

Source: MCP Registry · repository. Synced Oct 21, 2026, 17:53 UTC.

CompletionKit MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the CompletionKit MCP server safe?

No scan can promise that. What its definitions show: 54 tools, 36 of which can change, send or delete something, and 5 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the CompletionKit MCP server delete data?

Yes. 8 tools can delete, cancel or revoke something: prompts_delete, runs_delete, datasets_delete, metrics_delete, metric_groups_delete, metric_versions_dismiss and 2 more.

Can the CompletionKit MCP server do things that cannot be undone?

Yes. 2 tools can send, pay, publish or merge: prompts_publish and metric_versions_publish.

What tools does the CompletionKit MCP server have?

54 tools: 8 destructive, 2 irreversible, 26 write, 18 read. prompts_delete, runs_delete, datasets_delete, metrics_delete, metric_groups_delete, metric_versions_dismiss, provider_credentials_delete, tags_delete and 46 more.

Does the CompletionKit MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the CompletionKit MCP server send data to outside services?

3 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): prompts_publish, datasets_create_from_url and metric_versions_publish.

Has the CompletionKit MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 10, 2026.