09e99a09c6
21 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Hongming Wang
|
09e99a09c6 |
feat(a2a-mcp): add chat_history tool for prior turns with a peer
When a peer_agent push lands and the agent needs context from prior
turns with that workspace ("what task did this peer assign me last
hour?", "what did I tell them?"), the only options today are
re-deriving from memory (lossy) or scrolling activity_logs in the
canvas (no agent-facing tool). Surface the platform's existing
audit log directly via a new MCP tool so agents can read both sides
of an A2A conversation in chronological order.
Implementation:
- a2a_tools.py: new tool_chat_history(peer_id, limit=20, before_ts="")
hits /workspaces/<self>/activity?peer_id=X&limit=N (the new server
filter from molecule-core#2472). Reverses the DESC response into
chronological order so the agent reads top-down. Graceful error
envelope on validation/network/non-200 — never crashes the MCP
server, agent can branch on Error: prefix.
- platform_tools/registry.py: ToolSpec wired into the A2A section so
the rendered system-prompt block automatically includes it. Same
pattern as the existing inbox_peek/inbox_pop/wait_for_message.
- a2a_mcp_server.py: dispatch in handle_tool_call.
- executor_helpers.py: _CLI_A2A_COMMAND_KEYWORDS gets a None entry
(CLI runtimes don't expose chat history today; flip to a keyword
when a2a_cli grows a `history` subcommand).
- snapshots/a2a_instructions_mcp.txt regenerated.
Tests: 10 new branches in TestChatHistory (validation / param
forwarding / limit cap / before_ts pass-through / DESC→chronological
reorder / 400 verbatim / 500 generic / network exc / non-list resp).
Mutation-verified: reverting a2a_tools.py fails 10/10. Full test
suite remains green at 1516 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
Hongming Wang
|
f96bb9f860 |
docs(mcp): tagged server:NAME form in dev-channels reference
Claude Code 2.1.x's --dangerously-load-development-channels takes an allowlist of tagged entries (`server:<name>` or `plugin:<name>@<marketplace>`), not a bare switch. The instructions field's push-only-mode message and the inline comment in `_poll_timeout_secs` both referenced the old bare form. Update both so an agent or operator reading them lands on the right invocation — matched against the docs change in [molecule-docs PR #110](https://github.com/Molecule-AI/docs/pull/110). No behavior change (string-only edits in instructions text + comment). 33/33 tests still pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
f80e054a95
|
Merge pull request #2466 from Molecule-AI/feat/universal-push-via-instructions
feat(mcp): universal inbound delivery — instructions-driven polling + optional push |
||
|
Hongming Wang
|
c61a6ff9bd |
chore(mcp): drop unused module-level _CHANNEL_INSTRUCTIONS
The frozen copy was a self-justification — the comment claimed "tests + tooling rely on import-time identity" but no test or tooling code path actually references the binding. _build_initialize_result() calls _build_channel_instructions() fresh per call so env changes take effect, which is the documented runtime contract. github-code-quality flagged it; resolving the unused-variable thread so the staging branch protection's all-conversations-resolved gate clears. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
ea206043d8 |
feat(mcp): universal inbound delivery — instructions-driven polling + optional push
Why this exists --------------- Live evidence on 2026-05-01 caught a regression latent in #46's "push-feel inbound" closure: standard `claude` launches without `--dangerously-load-development-channels` silently drop our `notifications/claude/channel` emissions, so canvas/peer messages sat in the wheel inbox and never reached the agent loop until manual `inbox_peek`. The flag is research-preview-only; non-Claude-Code MCP clients (Cursor, Cline, OpenCode, hermes-agent, codex) never receive the notification at all because the method namespace is Claude- specific. Push-only delivery shipped as the universal contract is not actually universal. What this changes ----------------- Adds a poll path that works on every spec-compliant MCP client. The `initialize` `instructions` field — read by every client and surfaced to the agent's system prompt automatically — now tells the agent to call `wait_for_message(timeout_secs=N)` at the start of every turn. Push remains as the strictly-better delivery for hosts that opt in (Claude Code with the dev flag or a future allowlist entry), but is no longer load-bearing. Both paths converge on the same `inbox_pop` ack so duplicate-delivery on a push+poll race is impossible: whoever surfaces the message to the agent first pops it, the other side returns empty. Operator knob ------------- `MOLECULE_MCP_POLL_TIMEOUT_SECS` controls per-turn poll blocking (default 2s). 0 disables polling for push-only Claude Code with the dev flag. Above 60 clamps to 60 — protects against an accidental five-minute stall per turn. Resolved fresh on every `initialize` so a relaunch with new env is enough; no wheel rebuild required. Tests ----- - structural pins on the new instructions: `wait_for_message` + `timeout_secs` named, both PUSH PATH / POLL PATH labels present - env-resolution: default fallback, garbage fallback, negative fallback, 60s clamp - operator override: `MOLECULE_MCP_POLL_TIMEOUT_SECS=7` reaches the agent's instructions string - timeout=0 toggles to push-only-mode messaging (no wait_for_message call asked of the agent) - existing pins on push path, reply tools, prompt-injection defense, meta attributes — all preserved Successor to #46. Closure milestone for this PR (per feedback_close_on_user_visible_not_merge.md): launched `claude` against the published wheel, sent a canvas message, observed the agent surfaces the message inline at the start of its next turn without me running `inbox_peek` — verified live before declaring done. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
a3a496bced |
test(mcp): pin inbox→stdout bridge end-to-end with three failure-mode tests
Closes the dynamic-coverage gap on the `notifications/claude/channel` push-UX bridge — until now we had static pins on the wire shape (_build_channel_notification) and the initialize handshake, but the threading + asyncio + stdout chain that ships notifications to the host was never exercised under realistic conditions. The three failure modes anticipated in #2444 §2 are each now pinned: test_inbox_bridge_emits_channel_notification_to_writer Drives a fake inbox event from a daemon thread, asserts the notification lands on a real os.pipe-backed asyncio writer with the correct JSON-RPC envelope. Catches: bridge wired up incorrectly (no-op _on_inbox_message), run_coroutine_threadsafe drift, _build_channel_notification call missing. test_inbox_bridge_swallows_closed_pipe_drain_error Closes the pipe's read end before firing, captures the concurrent.futures.Future that run_coroutine_threadsafe returns, asserts its exception() is None. Catches: narrowing the broad `except Exception` in _emit (e.g. to RuntimeError), or removing it. Without the swallow, the future carries a ConnectionResetError and the test fails with a clear message naming the regression. test_inbox_bridge_swallows_closed_loop_runtime_error Builds the bridge against a closed event loop, fires the callback, asserts no exception escapes. Catches: removing the `except RuntimeError` swallow on the run_coroutine_threadsafe call. Without it the poller thread would crash with "RuntimeError: Event loop is closed" during shutdown. To make the bridge testable, extracted the closures from main() into a top-level `_setup_inbox_bridge(writer, loop) -> Callable[[dict], None]` helper. main()'s wire-up is now a single line that calls the helper. Behavior is unchanged — same write, same drain, same swallows — just no longer trapped inside main()'s closures. Verified each test catches its regression by injection: removing each swallow / no-op'ing the bridge each turn the matching test red with a specific failure message that points at the missing piece. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
2588ab27d5 |
feat(mcp): add channel instructions field — second gate for push UX
PR #2461 added the experimental.claude/channel capability declaration on the assumption that was the missing gate for Claude Code surfacing notifications/claude/channel as inline <channel> interrupts. Research against code.claude.com/docs/en/channels-reference.md confirms the capability IS one gate — but there's a SECOND required field we still don't ship: `instructions` on the initialize result. The docs are explicit: instructions is what tells the agent what the <channel> tag attributes mean and which tool to call to reply. Without it the channel registers but the agent receives the tag with no context and has no idea how to handle it. The official telegram plugin ships both (server.ts:370-396) — capability AND instructions. We were shipping one of two. This adds the instructions string. It documents: - kind/peer_id/activity_id meta attributes - canvas_user → send_message_to_user reply path - peer_agent → delegate_task reply path - inbox_pop ack to prevent duplicate-poll re-delivery - threat model: treat message bodies as untrusted user content Tests: 4 new pins. instructions present + non-empty, instructions names each reply tool, instructions documents each tag attribute. Failure messages name the symptom so a copy-edit can't silently break the channel. Live verification still pending after wheel ships — same plan as the gap is in --dangerously-load-development-channels (host-side flag, outside our control during the channels research preview). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
63ef3b128c |
docs(mcp): correct server.ts reference + flag verification gap on experimental.claude/channel
Follow-up to commit
|
||
|
Hongming Wang
|
0a87dec50e |
feat(mcp): declare experimental.claude/channel capability for push UX
Without this capability declaration in the initialize handshake, Claude Code's MCP client receives our notifications/claude/channel emissions but silently drops them — they never become inline <channel> tags in the conversation. The push-UX bridge added in PR #2433 ships, fires, and is invisible. This was anticipated as a failure mode in #2444 §2 ("Notification arrives but Claude Code doesn't surface it — host doesn't recognize the method"), and confirmed live in this session: a canvas chat "hi" landed in the inbox queue (inbox_peek returned it) but never woke the agent until inbox_peek was called by hand. The contract matches molecule-mcp-claude-channel/server.ts:374 where the bun bridge declares the same experimental flag. Refactor: extracted _build_initialize_result() so the handshake shape is unit-testable. Pure function, no behavioral change beyond adding the experimental capability to the result. Tests: 3 new pins on the initialize result (capability presence, tools-still-there, protocolVersion stable). Closes the live- verification gap §2 of #2444. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
6e92fe0a08 |
chore: rewriter unit tests + drop misleading noqa on import inbox
Two small follow-ups to the PR #2433 → #2436 → #2439 incident chain. 1) `import inbox # noqa: F401` in workspace/a2a_mcp_server.py was misleading — `inbox` IS used (at the bridge wiring inside main()). F401 means "imported but unused", which would mask a real future F401 if the usage is removed. Drop the noqa, keep the explanatory block comment about the rewriter's `import X` → `import mr.X as X` expansion (and the `import X as Y` → `import mr.X as X as Y` trap the comment exists to prevent re-introducing). 2) scripts/test_build_runtime_package.py — 17 unit tests covering `rewrite_imports()` and `build_import_rewriter()` in scripts/build_runtime_package.py. Until now the function had zero coverage despite the entire wheel build depending on it. Tests pin: bare-import aliasing, dotted-import preservation, indented imports, from-imports (simple + dotted + multi-symbol + block), the `import X as Y` rejection added in PR #2436 (with comment- stripping + indented + comma-not-alias edge cases), allowlist anchoring (`a2a` ≠ `a2a_tools`), and end-to-end reproduction of the PR #2433 failing pattern + the #2436 fix pattern. 3) Wire scripts/test_*.py into CI by adding a second discover pass to test-ops-scripts.yml. Top-level scripts/ tests live alongside their target file (parallels the scripts/ops/ test layout); the existing scripts/ops/ pass keeps running because scripts/ops/ has no __init__.py so a single discover from scripts/ root doesn't recurse. Two passes is simpler than retrofitting namespace packages. Path filter widened from `scripts/ops/**` to `scripts/**` so PRs touching the build script trigger the new tests. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
0acdf3bb56 |
fix(wheel): import inbox without alias to dodge rewriter collision
PR #2433 (notifications/claude/channel) shipped 'import inbox as _inbox_module' inside a2a_mcp_server.py:main(). The build script's import rewriter expands plain 'import inbox' to 'import molecule_runtime.inbox as inbox', so the original source became 'import molecule_runtime.inbox as inbox as _inbox_module', which is invalid Python. Caught at the publish-runtime + PR-built-wheel-smoke gate (the SyntaxError trace is in run 25200422679). The wheel didn't ship to PyPI because publish-runtime's smoke-import step refused to install it, but staging is currently sitting on a broken-build commit until this fix-forward lands. Changes: - a2a_mcp_server.py: lift `import inbox` to top of file (rewriter produces clean `import molecule_runtime.inbox as inbox`), call inbox.set_notification_callback directly in main() - build_runtime_package.py: rewrite_imports() now raises ValueError when it sees 'import X as Y' for any X in the workspace allowlist, instead of silently producing a syntax-error wheel. Operator gets a clear actionable error at build time pointing at the offending line + suggested rewrites ('from X import …' or plain 'import X'). The build-time gate (this PR's rewriter check) catches the regression class earlier than the smoke-time gate (PR #2433's failure). Adding 'PR-built wheel + import smoke' to staging branch protection's required checks is filed separately so this class doesn't merge again. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
0a3ec53f34 |
feat(mcp): notifications/claude/channel for push-feel inbox UX
Adds a notification seam to the universal molecule-mcp wheel so push- notification-capable MCP hosts (Claude Code today; any compliant client tomorrow) get inbound A2A messages as conversation interrupts instead of having to poll wait_for_message / inbox_peek. Wire-up: - inbox.py: module-level _NOTIFICATION_CALLBACK + set_notification_callback() Fires from InboxState.record() AFTER lock release, with same dict shape inbox_peek returns. Best-effort — a raising callback never prevents the message from landing in the queue. - a2a_mcp_server.py: _build_channel_notification() pure helper + bridge wiring in main() that schedules notifications via asyncio.run_coroutine_threadsafe (poller is a daemon thread, MCP loop is asyncio). - Method name 'notifications/claude/channel' matches the contract documented in molecule-mcp-claude-channel/server.ts:509. - wheel_smoke.py: pin set_notification_callback as a published name, same regression class as the 0.1.16 main_sync incident. Pollers (wait_for_message / inbox_peek) keep working unchanged for runtimes without notification support. Tests: 6 new in test_inbox.py (callback fires once on record, dedupe short-circuits before fire, raising cb doesn't break inbox, set/clear semantics), 5 new in test_a2a_mcp_server.py (method name pin, content mapping, meta routing, no-id JSON-RPC notification spec, missing- field tolerance). All 59 combined tests pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
b47d4ceb00 |
feat(workspace-runtime): add inbox polling for standalone molecule-mcp path
The universal MCP server (a2a_mcp_server.py) was outbound-only — agents
in standalone runtimes (Claude Code, hermes, codex, etc.) could
delegate, list peers, and write memories, but never observed the
canvas-user or peer-agent messages addressed to them. This blocked
"constantly responding" loops without forcing operators back onto a
runtime-specific channel plugin.
This PR closes the inbound gap with a poller-fed in-memory queue and
three new MCP tools:
- wait_for_message(timeout_secs?) — block until next message arrives
- inbox_peek(limit?) — list pending messages (non-destructive)
- inbox_pop(activity_id) — drop a handled message
A daemon thread polls /workspaces/:id/activity?type=a2a_receive every
5s, fills the queue from the cursor (since_id), and persists the cursor
to ${CONFIGS_DIR}/.mcp_inbox_cursor so a restart doesn't replay backlog.
On 410 (cursor pruned) we fall back to since_secs=600 for a bounded
recovery window. Activity-row → InboxMessage extraction mirrors the
molecule-mcp-claude-channel plugin's extractText (envelope shapes #1-3
+ summary fallback).
mcp_cli.main starts the poller alongside the existing register +
heartbeat threads. In-container runtimes (which have push delivery via
canvas WebSocket) skip activation, so inbox tools return an
informational "(inbox not enabled)" message instead of double-delivery.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
Hongming Wang
|
b54ceb799f |
fix: address 5-axis review findings on PR #2413
Critical:
- ExternalConnectModal.tsx: filledUniversalMcp substitution searched
for WORKSPACE_AUTH_TOKEN but the snippet's placeholder is now
MOLECULE_WORKSPACE_TOKEN (changed in the previous polish commit
|
||
|
Hongming Wang
|
169e284d57 |
feat(workspace-runtime): expose universal MCP server to runtime=external operators
Ship the baseline universal MCP path that any external runtime (Claude
Code, hermes, codex, anything that speaks MCP stdio) can use, before
optimizing per-runtime channels. Today the workspace MCP server only
spins up inside the container; external operators have no way to call
the 8 platform tools (delegate_task, list_peers, send_message_to_user,
commit_memory, etc.) from outside.
Three additive changes:
1. **`platform_auth.get_token()` env-var fallback** — adds
`MOLECULE_WORKSPACE_TOKEN` as a fallback when no
`${CONFIGS_DIR}/.auth_token` file exists. File-first preserves
in-container behavior unchanged. External operators (no /configs
volume) now have a way to supply the token without faking the
filesystem layout.
2. **`molecule-mcp` console script** — adds a new entry point in the
published `molecule-ai-workspace-runtime` PyPI wheel. Operators run
`pip install molecule-ai-workspace-runtime`, set 3 env vars
(WORKSPACE_ID, PLATFORM_URL, MOLECULE_WORKSPACE_TOKEN), and register
the binary in their agent's MCP config. `mcp_cli.main` is a thin
validator wrapper — it checks env BEFORE importing the heavy
`a2a_mcp_server` module so a misconfigured first-run gets a friendly
3-line error instead of a 20-line module-level RuntimeError
traceback.
3. **Wheel smoke gate** — extends `scripts/wheel_smoke.py` to assert
`cli_main` and `mcp_cli.main` are importable. Same regression class
as the 0.1.16 main_sync incident: a silent rename or unrewritten
import here would break every external operator on the next wheel
publish (memory: feedback_runtime_publish_pipeline_gates.md).
Test coverage:
- `tests/test_platform_auth.py` — 8 new tests for the env-var fallback:
file-priority, env-fallback, whitespace handling, cache, header
construction, empty-env-as-unset.
- `tests/test_mcp_cli.py` — 8 new tests for the validator: each
required var separately, file-or-env satisfies token requirement,
whitespace-only env treated as missing, help mentions canvas Tokens
tab.
- Full `workspace/tests/` suite green: 1346 passed, 1 skipped.
- Local end-to-end: built wheel, installed in venv, ran `molecule-mcp`
with no env → friendly error; with env → MCP server starts.
Why now / why this shape: user redirect was "support the baseline
first so all runtimes can use, then optimize". A claude-only MCP
channel leaves hermes/codex/third-party operators broken on
runtime=external. This PR ships the runtime-agnostic baseline; per-
runtime polish (claude-channel push delivery, hermes-native
bindings) is a follow-up PR. PR #2412 fixed the partner bug where
canvas Restart silently revoked the operator's token — the two
together unblock the external-runtime story end-to-end.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
Hongming Wang
|
f323def18f |
chore(build): include platform_tools in runtime wheel SUBPACKAGES
The PR-built wheel + import smoke gate refused the platform_tools
package because it's a new subdirectory under workspace/ that wasn't
in scripts/build_runtime_package.py:SUBPACKAGES. The drift gate (which
exists for exactly this reason) caught it cleanly:
error: SUBPACKAGES drifted from workspace/ subdirectories:
in workspace/ but NOT in SUBPACKAGES (will ship un-rewritten or
be excluded): ['platform_tools']
Adding platform_tools to SUBPACKAGES wires the package into the
runtime wheel + applies the canonical
from platform_tools.<x> -> from molecule_runtime.platform_tools.<x>
import-rewrite step that every other subpackage uses.
Verified locally: scripts/build_runtime_package.py succeeds, the
rewritten a2a_mcp_server.py reads
from molecule_runtime.platform_tools.registry import TOOLS
which matches the package layout in the wheel.
|
||
|
Hongming Wang
|
e9a59cda3b |
feat(platform): single-source-of-truth tool registry — adapters consume, no drift
Establishes workspace/platform_tools/registry.py as THE place tool
naming and docs live. Every consumer reads from it; nothing duplicates
the source. Closes the architectural gap behind the doc/tool drift
discussion 2026-04-28 — adding hundreds of future runtime SDK adapters
should not require touching tool names anywhere except the registry.
What the registry owns
ToolSpec dataclass with: name, short (one-line description), when_to_use
(multi-paragraph agent-facing usage guidance), input_schema (JSON Schema),
impl (the actual coroutine in a2a_tools.py), section ('a2a' | 'memory').
TOOLS list with 8 entries — delegate_task, delegate_task_async,
check_task_status, list_peers, get_workspace_info, send_message_to_user,
commit_memory, recall_memory.
What now reads from the registry
- workspace/a2a_mcp_server.py
The hardcoded TOOLS list (167 lines of hand-maintained dicts) is
gone. Replaced with a 6-line list comprehension over the registry.
MCP description = spec.short. inputSchema = spec.input_schema.
- workspace/executor_helpers.py
get_a2a_instructions(mcp=True) and get_hma_instructions() now
GENERATE the agent-facing system-prompt text from the registry.
Heading + per-tool bullet (spec.short) + per-tool when_to_use +
a section-specific footer. No more hand-maintained instruction
blocks that drift from reality.
- workspace/builtin_tools/delegation.py
Renamed delegate_to_workspace -> delegate_task_async to match
registry. check_delegation_status -> check_task_status. Added
sync delegate_task @tool wrapping a2a_tools.tool_delegate_task
(was missing for LangChain runtimes — CP review Issue 3).
- workspace/builtin_tools/memory.py
Renamed search_memory -> recall_memory to match registry.
- workspace/adapter_base.py, workspace/main.py
Bundle all 7 core tools (was 6) into all_tools / base_tools.
- workspace/coordinator.py, shared_runtime.py, policies/routing.py
Updated system-prompt-text references to use the registry names.
Structural alignment tests
workspace/tests/test_platform_tools.py — 9 tests pin every
registry-to-adapter mapping:
- registry names are unique
- a2a + memory partition is complete (no orphans)
- by_name lookup works
- MCP server registers exactly the registry's tool set
- MCP description equals registry.short for every tool
- MCP inputSchema equals registry.input_schema for every tool
- get_a2a_instructions text contains every a2a tool name
- get_hma_instructions text contains every memory tool name
- pre-rename names (delegate_to_workspace, search_memory,
check_delegation_status) cannot leak back
Adding a future tool means adding one ToolSpec; the test failure
list tells the author exactly which adapter to update.
Adapter pattern for future SDK support
When (e.g.) AutoGen or Pydantic AI gets adapters, the only work
needed for tool surfacing is "wrap registry.TOOLS in your SDK's
tool format." Names, descriptions, schemas, impl come from the
registry — adapter author writes zero strings.
Why this needed to ship now
PR #2237 (already in staging) injected MCP-world docs as the
default system-prompt content. Without the registry, those docs
said "delegate_task" while LangChain runtimes only had
"delegate_to_workspace" — workers see docs for tools that don't
exist (CP review Issue 1+3). PR #2239 was a tactical rename;
this PR is the structural fix that prevents the same class of
drift from recurring as new adapters ship.
PR #2239 was closed in favor of this — same renames, plus the
registry, plus structural tests. Single coherent change.
Tests: 1232 pass, 2 xfailed (pre-existing). 9 new in
test_platform_tools.py; 4 alignment tests in test_prompt.py from
#2237 still pass; original test_executor_helpers tests adapted to
the registry-driven world.
Refs: CP review Issues 1, 2, 3, 5; project memory
project_runtime_native_pluggable.md (platform owns A2A);
project memory feedback_doc_tool_alignment.md (this is the structural
fix for the tactical lesson).
|
||
|
Hongming Wang
|
7504aba934 |
feat(tools): tighten send_message_to_user description to forbid pasting URLs in body
Root-cause fix for #118 (chat attachments rendering as plain text links instead of download chips). User flagged with screenshot 2026-04-26 showing the Design Director agent pasting https://files.catbox.moe/… in the message body — chat rendered the URL as plain markdown text, unclickable in the canvas's bubble layout, and unreachable in any SaaS deployment where the user's browser can't egress to catbox. The structured `attachments` field already exists, the canvas's AttachmentChip already renders well, the WebSocket broadcast already carries attachments verbatim — the missing piece was the LLM choosing the body over the structured field. Tighten the tool description so it trains the right behavior. Three targeted strengthenings: 1. Top-level tool description: enumerated use case (4) now reads "via the `attachments` field (NEVER paste file URLs in `message`)". The all-caps NEVER + the explicit field name move the LLM toward the structured path on first read. 2. `message` param: adds an explicit DO NOT rule with rationale. Includes the SaaS-reachability reason so operators can grep for "SaaS" and find this design constraint instead of re-discovering it after a tenant complaint. Calls out catbox.moe + file:// by name as concrete examples of forbidden hosts (those are the two we've seen in production). 3. `attachments` param: leads with REQUIRED, lists the bad alternatives explicitly (pasting URLs, base64-encoding, telling user to look at a path). LLMs handle "use X, NOT Y" framings better than "use X" alone — observed during prompt-engineering iteration on hermes' tool descriptions. Tests pin all three load-bearing phrases (4 new in test_a2a_mcp_server.py) so a future doc edit that softens or drops them fails CI. Brittle by design — these are prompt-engineering invariants, not implementation details. This is the root-cause fix. A defensive canvas-side backstop (auto- detect download-shaped URLs in body and convert to chips) is a follow-up that could land separately if the steering proves insufficient in practice. Verification: - 1190/1190 workspace pytest pass - 4 new test_a2a_mcp_server.py cases all green Closes the steering half of #118. The structured-attachments-only contract was already enforced server-side (PR #2130 added per-attachment validation); this PR closes the prompt-side gap. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
Hongming Wang
|
af664e3e87 |
feat(tools): borrow hermes-style discipline — error/summary caps + sharper MCP descriptions
Three small wins from the hermes-agent design survey, bundled because
each is too small for its own PR but they all improve the priority
adapters (claude-code + hermes) immediately.
1. Hermes-style cap on telemetry fields, applied INSIDE report_activity
so every caller benefits without remembering. error_detail capped at
4096 (hermes' value); summary capped at 256 (one-liner ceiling). The
existing call site in tool_delegate_task already truncated error_detail
at 4096, but moving the cap into the helper closes the door on a
future caller pasting a giant traceback. response_text is NOT capped
(it's the agent's user-visible reply; truncating would silently drop
content). Pinned by 4 new tests including a negative-pin that
response_text MUST stay untruncated.
2. Sharper MCP tool descriptions for commit_memory + recall_memory —
hermes' delegate_task description literally says "WAIT for the response"
and delegate_task_async says "Returns immediately." LLMs pick the
right tool variant from descriptions; ambiguity costs accuracy.
- commit_memory now states it APPENDS (each call creates a row, no
overwrite) and that GLOBAL requires tier 0.
- recall_memory now states it's case-insensitive substring search
with no pagination, returns all matches, and that empty-query is
cheap and safer than a narrow keyword.
3. (no code change) Filed task #120 for the bigger user-flow win — a
per-workspace tool enable/disable menu in Canvas Config — and task
#121 for model-string passthrough (depends on #87 universal-runtime
refactor).
Verification:
- 1312/1312 Python pytest pass (was 1308, +4 new)
See task #119 for the architectural follow-ups (event-log layer,
declarative skill compat, observability config block) and project
memory `project_runtime_native_pluggable.md`.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
|
Hongming Wang
|
d028fe19ff |
feat(notify): agent → user file attachments via send_message_to_user
Closes the gap where the Director would say "ZIP is ready at /tmp/foo.zip"
in plain text instead of attaching a download chip — the runtime literally
had no API for outbound file attachments. The canvas + platform's
chat-uploads infrastructure already supported the inbound (user → agent)
direction (commit
|
||
|
Hongming Wang
|
479a027e4b |
chore: open-source restructure — rename dirs, remove internal files, scrub secrets
Renames: - platform/ → workspace-server/ (Go module path stays as "platform" for external dep compat — will update after plugin module republish) - workspace-template/ → workspace/ Removed (moved to separate repos or deleted): - PLAN.md — internal roadmap (move to private project board) - HANDOFF.md, AGENTS.md — one-time internal session docs - .claude/ — gitignored entirely (local agent config) - infra/cloudflare-worker/ → Molecule-AI/molecule-tenant-proxy - org-templates/molecule-dev/ → standalone template repo - .mcp-eval/ → molecule-mcp-server repo - test-results/ — ephemeral, gitignored Security scrubbing: - Cloudflare account/zone/KV IDs → placeholders - Real EC2 IPs → <EC2_IP> in all docs - CF token prefix, Neon project ID, Fly app names → redacted - Langfuse dev credentials → parameterized - Personal runner username/machine name → generic Community files: - CONTRIBUTING.md — build, test, branch conventions - CODE_OF_CONDUCT.md — Contributor Covenant 2.1 All Dockerfiles, CI workflows, docker-compose, railway.toml, render.yaml, README, CLAUDE.md updated for new directory names. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |