Attention

The vendor ships and maintains the server itself. A wrapper built by Zapier, Composio or a similar third party does not count as official. The status was established by hand on 2026-08-24. On 2026-09-04 the recorded URL served a documentation page, not an MCP endpoint. That is where to read about the server, not where to connect to it. An agent needs the second.

mcp_status, verbatim from the file:

official

mcp_url, verbatim from the file:

https://docs.attention.com/mcp/overview (re-verified 200 on 2026-08-28; corrected that day off a dead receipt, see notes. An unofficial community fork also exists: https://github.com/highgravitas/attention-mcp)

• https://docs.attention.com/mcp/overview
• https://github.com/highgravitas/attention-mcp

A tool below is one the server NAMES. Nobody has called it. That is the same two tier honesty rule the rest of the directory runs on: a named tool is research, and BENCH-TESTED stays the only claim that anybody ran anything.

This list came from a repo the vendor does not own. Those are that author's tools for the vendor's API, not the vendor's own published surface, and the two must not be read as the same thing.

• chat_id
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• conversation_id
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• conversation_ids
  array
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• days_back
  integer
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• deal_id
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• detailed_transcript
  boolean
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• from_date
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• include_timestamps
  boolean
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• items
  array
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• limit
  integer
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• offset
  integer
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• owner_email
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• participant_email
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• prompt
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• query
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• scorecard_id
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• scorecard_item_ids
  array
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• size
  integer
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• summary
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• team_uuids
  array
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• to_date
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• user_email
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• user_uuid
  string
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface
• user_uuids
  array
  evidence: in a README table · calling it reads · from a third party repo, so these are that author's tools, not the vendor's published surface

141 of the 741 entries that record an official or community MCP server carry a harvested tool list. The other 600 are unmeasured, which is not the same as empty. Harvest last run 2026-09-27. Every name across every server is on the tools index.

No CLI found by the 2026-09-27 harvest across vendor docs, npm, PyPI, Homebrew and GitHub. That is a probe result, not proof of absence.

api_gate, verbatim from the file:

paid. No free plan; pricing runs roughly $100-$500/month per third-party trackers, with deeper API/security features reserved for a Contact-Sales Enterprise tier.

No documentation URL recorded.

727 of 1252 entries are in the same position. Blank is legal and it is published as blank.

Not measured. github_url, github_stars, github_last_commit and github_archived are null on every entry in this build.

The refresh rail specced in SPEC section 7.2 has not been run. An unstamped star count is a lie, so nothing is shown rather than something stale.

A github.com URL already appears somewhere in this entry, which is a seed for that rail and not a measurement of repo health:

• https://github.com/highgravitas/attention-mcp

No GitHub organisation could be tied to attention.com with evidence on 2026-09-08.

Recorded by the harvest: github candidates were seen but none passed the evidence rules.

6 candidate accounts seen and rejected by the evidence rules: highgravitas, team-attention, AttentionX, buckaroo-it, attentiontech. A name match alone is never accepted; the account has to point at the vendor domain.

• Fetch a call transcript
• Summarize a meeting
• Extract deal signals from calls
• Write CRM records
• Score rep performance

A job tag means the vendor says the tool does this. It is not a test result, not proof the capability is reachable through the tool's MCP server, and not proof it is available on the gate this entry records.

Tagged by machine-pass on 2026-08-25 against the closed 56 job vocabulary. 271 of 1,252 entries carry at least one tag; 849 tags are assigned in total.

• https://docs.attention.com/mcp/overview
• https://docs.attention.com/mcp/authentication
• https://docs.attention.com/llms.txt
• https://www.attention.com/
• https://coldiq.com/tools/attention
• https://github.com/highgravitas/attention-mcp

6 source URLs. Raw sources field, verbatim:

https://docs.attention.com/mcp/overview, https://docs.attention.com/mcp/authentication, https://docs.attention.com/llms.txt, https://www.attention.com/, https://coldiq.com/tools/attention, https://github.com/highgravitas/attention-mcp

Every field above is rendered from directory.json exactly as the build produced it. Nothing is summarised and nothing is dropped. The one change made at render time is typographic and it is disclosed on the methodology page.