CHAP Coordinator
NPM · @BRIGHTBEAMAI/CHAP-COORDINATOR-MCP
Auditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations.
Available components
Last scanned (12 hours ago)
Scores higher than 42% of npm servers
Category Security
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security98
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- No install/post-install scripts declared.Pass
- 52 of 128 dependencies flagged as unhealthy (1 deprecated). View diagnostics → Partial
Provenance & Transparency48
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (Apache-2.0).Pass
- Actively maintained (last published 12 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability60
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 2902 tokens (~263/item across 11 items; 11 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management26
- Stability check failed: the tool surface changed between 0.3.3 and 0.3.5: 29 tool removals, 0 breaking changes, 1 additions. See how to fix → Fail
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 11 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 12 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a current MCP spec version (2026-07-28).Pass
Checks we ran (16)
- Malware scan
- Known CVEs
- Install scripts
- Dependency health
- Source repository
- Provenance
- License
- Maintenance
- Security disclosure
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
Does not apply to this package (2)
- Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
- Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
The score does not measure everything. What the score does not measure →
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 13 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 19 to 22.
- 11 Nov 26 +4
- Source repository: unverified → pass ▲ security
- 10 Nov 26 −5
- Provenance: pass → unverified ▼ security
- 9 Nov 26 +15
- Malware scan: unverified → pass ▲ security
- 8 Nov 26 −27
- Malware scan: pass → unverified ▼ security
- Stability: pass → fail ▼ security
- Schema quality: 229 → 263 ▼ functional
- Schema quality: excellent → good functional
- Destructive annotations: 100 → pass functional
- Package version: 0.3.4 → 0.3.5 functional
- 5 Nov 26 +1
- Stability: 0.97 → pass security
- 3 Nov 26 −1
- Stability: pass → 0.93 functional
- 30 Oct 26 +15
- Malware scan: unverified → pass ▲ security
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 21 Nov 2026 · Analysed npm/@brightbeamai/chap-coordinator-mcp@0.3.5
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Dependencies 128 packages
| Packages resolved | 128 |
|---|---|
| Deprecated | 1 |
| Stale | 51 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
chap.audit.read Read audit log ~210
Read entries from a workspace's audit log, optionally within a sequence range and filtered by method, sender, task or outcome. This is the query surface for everything the workspace has recorded, decisions and overrides included. An accepted call is under 'envelope'. A refused attempt the coordinator recorded, such as a decision by someone the review was not addressed to, is under 'request' with an 'outcome' giving the error code, and it did not take effect. Reads only: it records nothing, so it never appears in its own output. Entries come back as recorded, without aggregation, so grouping them, by tag for instance, is the reader's job.
| Name | Type | Req | Description |
|---|---|---|---|
| filter | object | – | Narrows the entries returned. Conditions combine with AND, and are applied within the sequence window rather than before it. |
| range | object | – | Sequence window to return. Omit for the whole log. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.audit.verify_chain Verify audit chain ~129
Replay a workspace's prev-hash chain. Only status verified with ok true means the log was checked and is intact. Status not_evaluated with ok false means part of the log was never checked, so its integrity is unknown and must not be reported as verified; entries_unchecked says how much. An error means the chain is broken or absent.
| Name | Type | Req | Description |
|---|---|---|---|
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.decide.approve Approve ~279
Approve the artefact under review as it stands. Only a reviewer the review was addressed to may decide, and anyone else is refused with -32011. The task completes once the review's rule is satisfied, so under all_approve or quorum:N an approval may leave the review open awaiting others. Use chap.decide.override instead to accept a corrected version.
| Name | Type | Req | Description |
|---|---|---|---|
| approved_artefact_digest | string | – | Optional. SHA-256 over the JCS canonicalisation of the artefact under review, in the form `sha256:<hex>`. When present it binds the decision to the exact content reviewed, and a mismatch is refused w… |
| comment | string | – | The reviewer's note on this decision. Recorded in the audit entry for the decision. |
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| tags | array | – | Workspace-defined labels for this decision, e.g. ['tone', 'unsupported-claim']. Recorded in the audit entry. chap.audit.read does not filter on tags, so grouping by tag is done by the reader. |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.decide.override Override with a correction ~416
Correct the artefact under review with an RFC 6902 JSON Patch and accept the result. The patch, the rationale and any tags are recorded together, so the audit log holds what was changed and why, rather than only that the work was not accepted as written.
| Name | Type | Req | Description |
|---|---|---|---|
| approved_artefact_digest | string | – | Optional. SHA-256 over the JCS canonicalisation of the artefact under review, in the form `sha256:<hex>`. When present it binds the decision to the exact content reviewed, and a mismatch is refused w… |
| diff | array | yes | RFC 6902 JSON Patch operations applied to the artefact under review. The patched artefact becomes the task output; a patch that does not apply is refused with -32012. |
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| intent_preserved | boolean | – | True when the edit refines the decision the draft was making, false when it substitutes a different one. Recorded in the audit entry. |
| logical_id | string | – | Caller-chosen identifier for the item being decided, stable across revisions and overrides of the same underlying content. Recorded in the audit entry. |
| policy_refs | array | – | Identifiers of the policies or guidelines this correction applies, e.g. ['policy:no-delivery-promises']. Recorded in the audit entry. |
| rationale | string | yes | Why the correction was made. Required on every override and recorded in the audit entry, since the diff shows what changed but not why. |
| tags | array | – | Workspace-defined labels for this decision, e.g. ['tone', 'unsupported-claim']. Recorded in the audit entry. chap.audit.read does not filter on tags, so grouping by tag is done by the reader. |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.decide.reject Reject ~304
Reject the artefact under review. The task is declined, or returns to in_progress when request_revision is set, which is the choice between ending the work and sending it back. Only a reviewer the review was addressed to may decide. Use chap.decide.override instead to correct the artefact and accept it rather than refuse it.
| Name | Type | Req | Description |
|---|---|---|---|
| approved_artefact_digest | string | – | Optional. SHA-256 over the JCS canonicalisation of the artefact under review, in the form `sha256:<hex>`. When present it binds the decision to the exact content reviewed, and a mismatch is refused w… |
| comment | string | – | The reviewer's note on this decision. Recorded in the audit entry for the decision. |
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| request_revision | boolean | – | When true the task returns to 'in_progress' rather than 'declined', so the assignee can revise and resubmit. |
| tags | array | – | Workspace-defined labels for this decision, e.g. ['tone', 'unsupported-claim']. Recorded in the audit entry. chap.audit.read does not filter on tags, so grouping by tag is done by the reader. |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.review.acknowledge Acknowledge review ~197
Say that this reviewer has begun the open review on a task, before deciding it. The call is recorded, so the audit log shows when each reviewer took the review up as well as when they decided. It decides nothing: the review stays open and the acknowledgement counts towards no rule. Only someone who may decide the review may send it, and anyone else is refused with -32011. A task with no open review is refused with -32010, and a second acknowledgement from the same reviewer in the same round with -32602. Returns requested_at, the time the round opened, which names it.
| Name | Type | Req | Description |
|---|---|---|---|
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.task.complete Submit task output ~277
Submit a task's output. A task that requires review does not complete: the output is held as the artefact under review, the task moves to review_requested, and a reviewer decision completes it. Any other task completes immediately.
| Name | Type | Req | Description |
|---|---|---|---|
| confidence | string | – | Self-reported confidence in the output, between 0 and 1. Written as a decimal string, e.g. "0.86". CHAP canonicalisation accepts integers only, so a JSON number with a fractional part is refused with… |
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| output | – | – | The task's output artefact. Pass a JSON object or array. A JSON-encoded string is parsed back to the structured value before dispatch, so patches in chap.decide.override apply against a real object. |
| routing_hints | object | – | Signals recorded on the task and read by the routing/1.0 methods: task.route, review.depth and escalate.auto. Recording a hint has no effect on its own; it is consulted only when one of those methods… |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.task.create Create task ~463
Create a task: a unit of work assigned to one participant. Set review_required to make completion depend on a reviewer decision rather than on the assignee. Returns a task_id that the rest of the lifecycle takes. Supplying idempotency_key is what makes a retry safe: a repeat carrying a key already seen returns the original task, while a repeat without one creates a second task. Assigning to a paused member is refused.
| Name | Type | Req | Description |
|---|---|---|---|
| assignee | string | yes | Who the task is assigned to. Must be a workspace member, and must not be paused: assigning to a paused member is refused with -32063. |
| deadline | string | – | When the task is due, as an ISO 8601 timestamp. Recorded on the task; the coordinator does not act on it. |
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| idempotency_key | string | – | Caller-chosen key for safe retries. A second create carrying a key already seen in this workspace returns the original task and records nothing further. The workspace retains the 10,000 most recent k… |
| input | object | yes | Task-specific input payload. |
| kind | string | yes | Task kind, e.g. 'draft_response' or 'review'. Free text; the coordinator records it without interpreting it. |
| mode | string | – | Mode for this task, defaulting to the workspace mode. A mode above the workspace ceiling is refused with -32040. |
| review_required | boolean | – | When true, chap.task.complete opens a review instead of completing: the output becomes the artefact under review, and the task reaches 'completed' only on a reviewer decision. chap.task.update cannot… |
| routing_hints | object | – | Signals recorded on the task and read by the routing/1.0 methods: task.route, review.depth and escalate.auto. Recording a hint has no effect on its own; it is consulted only when one of those methods… |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.task.update Change task state ~247
Move a task to a new state. Only the transitions in the specification's lifecycle table are accepted, and a task that requires review cannot be completed here. A task whose review no decision has closed, or a successor that carries one on, is not declined (-32602).
| Name | Type | Req | Description |
|---|---|---|---|
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| progress_note | string | – | Short note on what changed, kept in the task's history. |
| state | string | yes | The state to move the task to. Only the transitions in SPECIFICATION.md 8.1 are legal from the task's current state; others are refused with -32602. A review opens with chap.review.request, or with c… |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.whisper.ask Ask a whisper ~279
Put one question to one or more participants, with a deadline and a default. If the deadline passes unanswered the default applies, so a task is never blocked waiting on a reply.
| Name | Type | Req | Description |
|---|---|---|---|
| deadline_ms | integer | yes | How long the whisper stays open, in milliseconds from now. Once it passes, the whisper lapses and default_if_lapsed is applied. |
| default_if_lapsed | – | yes | The value applied if the deadline passes with no answer. Required, so that a lapsed whisper still has a defined outcome. |
| from | string | – | This server acts as agent:assistant: it fills this in, and refuses a call that names anyone else. |
| options | array | – | Multiple-choice options. When present, an answer must name one of these ids in answer_option; any other id is refused with -32022. |
| question | string | yes | The question being put. A whisper is answered on its own, without the recipient opening the task. |
| task_id | string | yes | Task identifier returned by chap.task.create. |
| to | array | yes | Who is being asked. |
| urgency | string | – | How urgent the question is. Recorded and passed on to the client; it does not change the deadline or the lapse behaviour. |
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
chap.workspace.describe Describe workspace ~52
Report the current state of a workspace: its members, enabled profiles, audit length, and task and override counts.
| Name | Type | Req | Description |
|---|---|---|---|
| workspace | string | yes | Workspace identifier, e.g. 'wsp_techcorp_support'. |
No output schema declared.
No examples provided.
How do I install the CHAP Coordinator MCP server?
CHAP Coordinator runs locally as an npm package, launched with npx -y @brightbeamai/chap-coordinator-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · @brightbeamai/chap-coordinator-mcp
claude mcp add brightbeamai-chap -- npx -y @brightbeamai/chap-coordinator-mcp
What is the CHAP Coordinator MCP server?
CHAP Coordinator is an MCP server listed in the public MCP registry as io.github.BrightbeamAI/chap. Auditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations. This page covers its npm package (@brightbeamai/chap-coordinator-mcp).
Is the CHAP Coordinator MCP server safe to use?
CHAP Coordinator scores 71 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 21 November 2026. It declares no install or post-install scripts. 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 CHAP Coordinator MCP server expose?
CHAP Coordinator exposes 11 tools: chap.workspace.describe, chap.task.create, chap.task.update, chap.task.complete, chap.audit.read, and 6 more. Their descriptions and schemas cost roughly 2,853 tokens of context every time the server is loaded.
Is the CHAP Coordinator MCP server still maintained?
CHAP Coordinator 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 licence is the CHAP Coordinator MCP server under?
CHAP Coordinator declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.
What did VerifyMCP test on the CHAP Coordinator MCP server?
VerifyMCP ran 16 checks on CHAP Coordinator, covering Malware scan, Known CVEs, Install scripts, Dependency health and 12 more. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.