Skip to content
DashboardWebsite

Platform MCP

The platform MCP lets an agent (Claude Code, Cursor, VS Code, Codex, or your own agent) do what you can do in the dashboard: run and triage analyses, manage rules, products and repositories, administer your organization and, for a platform admin, the platform. It acts as you, never with more rights.

It is the only MCP endpoint. A CI job connects with an org API key that has the pipeline role: it sees the analysis tools and nothing else.


WhereURL
Its own hosthttps://mcp.<your-domain>/
On the API hosthttps://api.<your-domain>/platform-mcp/
Local stackhttp://localhost:8080/platform-mcp/

In the dashboard: Account → Agent access. Create an access token for yourself, then:

Terminal window
claude mcp add --transport http opentremor https://mcp.example.com/ \
--header "Authorization: Bearer ot_pat_..."

Any MCP client that takes a JSON server list:

{
"mcpServers": {
"opentremor": {
"url": "https://mcp.example.com/",
"headers": { "Authorization": "Bearer ot_pat_..." }
}
}
}

An org API key (X-API-Key) works too, for CI or a shared bot that acts in one organization:

Terminal window
claude mcp add --transport http opentremor https://mcp.example.com/ \
--header "X-API-Key: $OPENTREMOR_API_KEY"

Claude.ai and ChatGPT connectors take only the URL. Add https://mcp.<your-domain>/ as a custom connector: the client opens an OpenTremor page where you sign in, choose the scope and, optionally, one organization, and allow it. Disconnect it in the dashboard under Account → Agent access → Connected apps. See Access Tokens for Agents.


The agent sees only the tools your credential allows: an access token’s scope and your role in the organization decide the list. A read token gets read tools only; the platform tools appear only for a platform admin’s admin token.

Tools that take org_id act in the organization you specify. If you omit it, they use the organization in the caller’s current context. For access tokens, you can pass org_id to act in another organization you belong to. The MCP server forwards an explicit org_id to the API as X-OpenTremor-Org.

Tools marked destructive take confirm: true. Without it they describe what they would do, so the agent can ask you first.

Some actions stay in the dashboard: writing SSO, SCIM or sign-in methods (a mistake locks everyone out), deleting an organization, granting platform admin, and anything about your own account (password, 2FA, tokens).


The agent runs the client-led loop with its own model:

  1. ingest splits the input into units and returns a receipt with the revision_id.
  2. claim_units hands out a batch of units (5 by default, 20 at most) with their rules and the contract that says how to write an analysis. Its status says what to do next: units (analyse and submit them), waiting (another caller holds the remaining units; claim again after retry_after_seconds) or done.
  3. submit_analyses stores the analyses of several units in one call and reports each unit’s result: stored, rejected with errors to fix, not_in_revision, or claimed_by_other.
  4. get_report returns the report once the status is done.

Each ruleset and the contract are sent once: a unit names its ruleset_id, and the agent lists the ids it already holds in known_rulesets and known_contract. An agent that stops early calls release_units, so the units it claimed do not wait out their 10-minute claim.

The server also offers:

KindNameWhat it is
Promptanalyse_changesetThe loop above, as instructions for the agent
Resourceopentremor://packs/{pack}/rulesA pack’s ruleset in your current organization

ToolKindNeedsWhat it does
whoamiread—Who you are acting as: user, organization, role, permissions, and whether through an access token
list_my_orgsread—The organizations you belong to, with your role in each, under items
list_packsreadanalysis:readThe packs enabled for the organization: the kinds of input you can analyse
get_pack_rulesreadanalysis:readA pack’s ruleset in Markdown, under text, to preview it
ingestwriteanalysis:runSplit raw input (a plan, a diff) into units under changeset
claim_unitswriteanalysis:runClaim up to limit units (1 to 20) of changeset to analyse
submit_analyseswriteanalysis:runStore the analyses of claimed units, one entry per unit, in one call
release_unitswriteanalysis:runHand back the units you claimed in changeset and will not submit, so they are claimable at once
start_server_analysiswriteanalysis:runLet the server call the LLM and analyse raw_input in one step, with one of the org’s LLM connections
list_changesetsreadanalysis:readThe organization’s changesets, newest first, each with its repository, that repository’s product, its pull request and its open findings by severity
list_changeset_revisionsreadanalysis:readA changeset’s revisions, newest first: one per push or analysis
get_revision_changesreadanalysis:readThe findings one revision opened, closed and left open, compared with the revision before it
get_reportreadanalysis:readThe report of a changeset: its current revision, or revision_id
get_merge_gatereadanalysis:readWhether a changeset’s findings allow its change to merge, and why
list_findingsreadfindings:readFindings for the organization, filtered
findings_summaryreadfindings:summaryCounts of the organization’s findings by status and severity
triage_findingwritefindings:triageSet a finding’s triage status, with a note explaining why
submit_feedbackwritefeedback:writeRecord a verdict on a finding
list_rulesreadrules:readThe organization’s custom rules
get_rulereadrules:readOne custom rule
get_rule_coveragereadrules:readWhere a custom rule runs: for each repository, runs or the first reason it does not. The reasons, in the order they are checked: rule_disabled, pack_off, other_variant, out_of_scope, excluded
reset_rulewriterules:manageRestore a built-in rule’s pack wording, as shown by pack_update on a rule the organization edited
create_rulewriterules:manageAdd a custom rule
update_rulewriterules:manageChange a custom rule
delete_ruledestructiverules:manageDelete a custom rule
list_rule_categoriesreadrule_categories:readThe organization’s rule categories, to file a custom rule under
list_rule_proposalsreadfeedback:readRule changes suggested by review feedback or by the organization’s ADRs
review_rule_proposalwriterule_proposals:manageApply, defer or dismiss a rule proposal
rule_healthreadfeedback:readHow often reviewers agree with each rule: which rules are noisy
list_productsreadproducts:readThe organization’s products: groups of repositories owned by teams
get_product_overviewreadfindings:summaryA product’s repositories, each with its open findings by severity, its number of changesets and its last revision, and the product’s totals
get_product_reportreadfindings:readA product’s open findings: one section per repository with its recent changesets, after the rules with the most findings
get_repository_reportreadfindings:readA repository’s open findings and recent changesets
create_productwriteproducts:manageAdd a product, optionally owned by some teams and holding some repositories
list_goalsreadproducts:readWhere a product is heading: its goals, each with ordered steps and their status
create_goalwritegoals:manageAdd a goal to a product, with its steps in order
get_goalreadproducts:readOne goal with its steps, and the pull requests that advanced or regressed it, newest first
update_goal_stepwritegoals:manageSet a goal step’s status: todo, in_progress, done or skipped
list_repositoriesreadrepositories:readThe organization’s repositories known to OpenTremor, filtered
update_repositorywriterepositories:manageChange a repository’s settings
resolve_pack_suggestionwriterepositories:manageAct on a pack in a repository’s suggested_packs
preview_repository_scanreadrepositories:scanWhat a scan of a repository’s existing code would read
start_repository_scanwriterepositories:scanScan a repository’s existing code on its default branch
get_repository_scanreadrepositories:readA repository’s last scan: its status, the slices it ran, those left to resume, its job, and the units it analysed and reused
list_org_packsreadpacks:manageWhich packs are enabled for the whole organization
set_pack_enabledwritepacks:manageEnable or disable a pack for the whole organization
list_jobsreadjobs:listThe organization’s analysis jobs, newest first
get_jobreadjobs:readOne job’s status and progress
cancel_jobwritejobs:runStop a queued or running job
get_orgreadorg:readThe organization’s details
list_membersreadmembers:readThe organization’s members and their roles
change_member_rolewritemembers:manageGive a member another role
remove_memberdestructivemembers:manageRemove a member from the organization
invite_memberwriteinvites:manageInvite someone by email
list_invitesreadinvites:manageThe organization’s open invitations
list_teamsreadteams:readThe organization’s teams
create_teamwriteteams:manageAdd a team
add_team_memberwriteteams:manageAdd a member of the organization to a team
list_rolesreadroles:readBuilt-in and custom roles, with the permissions each holds
list_api_keysreadapi_keys:manageThe organization’s API keys (service accounts), without their secrets
create_api_keywriteapi_keys:manageCreate an API key for a CI job or a shared bot
revoke_api_keydestructiveapi_keys:manageRevoke an API key
get_agent_accessreadagent_access:manageWhether members’ access tokens act in the organization
set_agent_accesswriteagent_access:manageAllow members’ access tokens, cap them at read, or refuse them in the organization
get_ssoreadsso:manageThe organization’s SSO connection, read-only: changing it stays in the dashboard
list_notification_channelsreadnotifications:manageWhere the organization’s notifications go
list_llm_connectionsreadllm_connections:readThe organization’s LLM connections, for start_server_analysis
get_usagereadbilling:readThe organization’s usage and spend this month
query_audit_logreadaudit:readThe organization’s audit log, newest first
admin_list_usersreadplatform adminEvery account on the platform, filtered by name, email or organization
admin_set_user_activewriteplatform adminActivate or deactivate an account across the platform
admin_list_orgsreadplatform adminEvery organization on the platform
admin_get_orgreadplatform adminOne organization, as seen by a platform admin
admin_set_org_planwriteplatform adminSet an organization’s plan (cloud deployments only)
admin_lock_orgwriteplatform adminLock or unlock an organization
admin_get_settingsreadplatform adminThe platform settings: each field’s value and whether it is overridden
admin_update_settingswriteplatform adminChange platform settings
admin_query_auditreadplatform adminThe platform-wide audit log, newest first

The service is OpenTremor-mcp. It is stateless, holds no secret, and calls the API with each caller’s own credential.

VariableDefaultMeaning
OPENTREMOR_API_URLhttp://localhost:8000Where the API is, as this service reaches it
PORT8090Port of the MCP endpoint
MCP_PATH/Path of the MCP endpoint
MCP_PUBLIC_URL—This service’s public URL. Together with the next one, it enables OAuth discovery
OPENTREMOR_PUBLIC_API_URL—The API’s public URL: the OAuth authorization server

GET /healthz answers without a credential. A request to the MCP endpoint without a credential gets 401. See OpenTremor-mcp#0001 for the design.