@anyforge/mcp-server 3.14.5 → 3.15.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +14 -5
- package/dist/index.js +8 -8
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -259,11 +259,11 @@ complete capability and local/remote workspace boundary.
|
|
|
259
259
|
| `anyforge_list_program_files` | read | A program's Files tab (`GET /programs/{id}/files`) |
|
|
260
260
|
| `anyforge_split_work_item` | write | Replace one unstarted work item with multiple children |
|
|
261
261
|
| `anyforge_merge_work_items` | write | Replace multiple unstarted work items with one merged item |
|
|
262
|
-
| `anyforge_list_crews` | read | Discover crews by status
|
|
262
|
+
| `anyforge_list_crews` | read | Discover crews by status and active state, newest first. Returns 25 rows unless you pass `limit` (up to 200) |
|
|
263
263
|
| `anyforge_get_crew` | read | Crew state + recent outputs + comments. Leads with status, halt reason, work item, PR and timestamps; `shape: "summary"` (default) excerpts long text and leaves out large snapshots, naming each; `messages: 0` is the status read; `shape: "full"` for every field |
|
|
264
264
|
| `anyforge_post_crew_comment` | write | Comment on / @-mention people on a crew (the crew does not read comments; use `anyforge_message_crew` to steer a running crew, `anyforge_amend_crew` for one at a gate) |
|
|
265
265
|
| `anyforge_message_crew` | approve | Message a running crew's Engineer ("skip X", "check Z first"). It reads the message before its next step and confirms in the crew chat; a crew not yet started or stopped inside the Engineer reads it on its first step or on resume. Refused with `crew_not_reading_messages` at a gate, in review or when finished. Optional `idempotencyKey` makes a retry safe |
|
|
266
|
-
| `anyforge_resume_crew` | approve | Resume a PAUSED/HALTED crew |
|
|
266
|
+
| `anyforge_resume_crew` | approve | Resume a PAUSED/HALTED crew. Optional `guidance` (up to 4000 characters) resumes with guidance: the Engineer's first resumed round runs with it, in the same call. Guidance replaces the automatic recovery of a consumed amend that a plain resume of a crew halted at its ship gate runs, rather than running with it. A crew the guidance cannot reach is refused with `not_resumable` and the guidance is not stored: one that is not PAUSED/HALTED, or one stopped at a gate (paused at any gate, or halted at the ship gate); use `anyforge_amend_crew` there. Requires `approve` scope. |
|
|
267
267
|
| `anyforge_halt_crew` | approve | Halt a crew |
|
|
268
268
|
| `anyforge_amend_crew` | approve | Continue & finish — send a crew at its HIL gate back to the engineer with guidance. Optional `postToPullRequest` also posts the notes on the pull request; left out, the organization setting decides (off by default) |
|
|
269
269
|
| `anyforge_fix_crew_ci` | approve | Fix CI: send a crew whose pull request has failing checks back to the Engineer to fix them on the same PR. Works from PR review (HIL_SHIP) or a HALTED/PAUSED crew that has a PR; refused when GitHub does not report CI as failing. Requires `approve` scope and spends crew tokens. Only call when the operator asks. |
|
|
@@ -276,6 +276,7 @@ complete capability and local/remote workspace boundary.
|
|
|
276
276
|
| `anyforge_get_repo_analysis` | read | One repo's coverage / architecture / gap summary |
|
|
277
277
|
| `anyforge_check_dispatch_readiness` | read | With `workItemId`, list every blocker that would stop a crew starting on that item; with only `repoFullName`, check the repository's analysis gate. Does not launch a crew |
|
|
278
278
|
| `anyforge_analyze_repos` | write | Start Phase-1 analysis for one or more repos; Phase 2 follows automatically when Phase 1 completes |
|
|
279
|
+
| `anyforge_request_pr_review` | write | Request the review `@anyforge review` produces for an existing pull request (`repo` as `owner/name`, `prNumber`). Returns when queued; needs an owner or operator who has connected GitHub. Example: `{ "repo": "acme/api", "prNumber": 42 }` |
|
|
279
280
|
| `anyforge_start_coverage_sweep` | write | Paced Phase 1→2 analysis sweep over a program/repo set |
|
|
280
281
|
| `anyforge_refresh_repo_index` | write | Refresh the repo index + committed-coverage detection now |
|
|
281
282
|
| `anyforge_upload_coverage` | write | Upload test-coverage measurements for one or many repos (e.g. synced from SonarQube) |
|
|
@@ -328,6 +329,9 @@ complete capability and local/remote workspace boundary.
|
|
|
328
329
|
| `anyforge_create_risk` | write | Record an accepted risk on a program, initiative or work item: owner (defaults to you), `reviewBy` date or `acceptedUntilMilestoneId`, trigger, mitigation. A risk never blocks dispatch, `Depends on` or the critical path |
|
|
329
330
|
| `anyforge_list_risks` | read | List accepted risks for a program (`programId`) or one subject (`subjectKind` + `subjectId`); closed risks only with `includeClosed` |
|
|
330
331
|
| `anyforge_update_risk` | write | Change a risk's status (for example `MATERIALISED` or `CLOSED`), owner, review date, mitigation or follow-up work items; only the fields passed change |
|
|
332
|
+
| `anyforge_list_connected_tools` | read | The company MCP servers an owner opened to local agents, and the tools on them you may use (`access` read or write, `needsApproval`, `actsAs` company or you). Pass `server` for input schemas. Needs an Owner, Operator or Editor on a personal key |
|
|
333
|
+
| `anyforge_call_connected_tool` | read | Call one listed tool (`server`, `tool`, `arguments`). A read runs with the stored credential and returns its result; a write, or a tool marked `needsApproval`, is raised for a person's approval and returns `pending_approval` with a `callId`. Writes need an Owner or Operator and the `write` scope. Every call is audited without its arguments or result |
|
|
334
|
+
| `anyforge_get_connected_tool_call` | read | Where your approval-gated call stands (`pending_approval`, `approved`, `executed`, `failed`, `rejected`, `expired`) and, once settled, what happened |
|
|
331
335
|
| `anyforge_list_report_filter_values` | read | The values a query filter field may take: an enum's closed set, or real ids with names (`query` looks one up by name); needs a role that edits reports |
|
|
332
336
|
| `anyforge_triage_ticket` | read | Run or continue a Support Triage interview — classification, readiness scoring, owning-team routing — and get a paste-ready rendering plus the structured draft; pass the same `sessionId` to continue an interview. Never files anything itself |
|
|
333
337
|
| `anyforge_list_triage_eval_cases` | read | List permitted triage eval cases; historical cases require Owner/Operator |
|
|
@@ -453,9 +457,14 @@ Every tool advertises 2025-06-18 MCP metadata so clients can present it well:
|
|
|
453
457
|
|
|
454
458
|
- **Title + behaviour annotations** — a human-readable `title` plus hints
|
|
455
459
|
(`readOnlyHint`, `destructiveHint`, `idempotentHint`, `openWorldHint`) derived
|
|
456
|
-
from
|
|
457
|
-
|
|
458
|
-
|
|
460
|
+
from each tool's policy, matching the explicit policies and Scope column above.
|
|
461
|
+
Following MCP, `destructiveHint` is false only for a write that just adds
|
|
462
|
+
something (posting a comment, attaching a file); every other write, creates
|
|
463
|
+
included, can change existing state and is marked destructive. `openWorldHint` is set
|
|
464
|
+
on tools that reach outside the workspace, such as a comment shared on a
|
|
465
|
+
client's report link or an invitation email. Clients use these to, e.g.,
|
|
466
|
+
auto-run read-only tools and confirm before a destructive one. They are
|
|
467
|
+
advisory only: the control API still re-checks the key's scopes on every
|
|
459
468
|
call, so annotations never widen access.
|
|
460
469
|
- **Structured output** — every tool declares an `outputSchema` (a JSON object)
|
|
461
470
|
and returns the result as `structuredContent` alongside the human-readable
|