RU
Agent teams, swarms, memory and hooks for Claude Code and Codex, served over MCP.
Nothing to run from here
Harflow is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
94/100
trust score
3
downloads a day
local
runs on your machine
435
downloads in 30d
What it does
Dispatch coding tasks to local headless agents with persistent history and workflows.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user harflow-mcp '--' uvx harflow-mcp
# 2 — ask your agent for something these tools do
# list_backends, start_task, list_approval_requests and 61 more
Harflow costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
HDAll 1 server
Harflow-Dev1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Harflow as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need uv (docs.astral.sh/uv): uvx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user harflow-mcp '--' uvx harflow-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
64 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Report the available agent backends, what is installed, and what each can actually do.
Takes no parameters.
Dispatch a coding task to a headless agent and return immediately.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| prompt | string | yes | Instructions for the agent. Be specific about the desired end state. Write it in plain prose: preserve spaces and word boundaries, and never concatenate words or strip whitespace to shrink the payload |
| repo_path | string | yes | Absolute path to a git repository; the agent's working directory. |
| backend | string | no | Which harness to use — "claude", "codex", "opencode", "vibe" or "antigravity". See list_backends. |
| freedom | string | no | "read_only", "write_in_repo" (default), "publish", or "unrestricted". "publish" sits between "write_in_repo" and "unrestricted": it authorizes an attempt to remote publishing, including commit/push/PR |
| model | string | no | Model for this run, in the backend's own naming. Rejected outright, rather than silently ignored, on a backend with no model-selection flag at all — vibe is the one such case; see list_backends' capab |
| max_turns | integer | no | Cap on agent turns. Only some backends support this; asking for it on one that does not is an error rather than being silently ignored. |
| background_idle_seconds | no | Optional finite positive live-input background wait bound, at most 21600 seconds. Capped/classic runs do not use live-input idle waiting. Continuations inherit the saved effective bound unless overrid | |
| reasoning_effort | string | no | "low", "medium", "high" or "xhigh", passed to the backend verbatim. Rejected up front when the chosen *backend* has no effort control at all (vibe is config-only and has none), or is asked for a level |
| network | no | Optional boolean asking for network access independently of freedom: True asks harflow_mcp to impose no network barrier of its own, False asks it to impose one, and None (the default) keeps each fre | |
| caller_approvals | no | Opt in to version-qualified exact-action caller decisions: Bash, plus Read on Claude Code 2.1.310 (approving a Read can disclose files outside the repository). Only uncapped non-native Claude Code wri | |
| interaction_owner | string | no | caller by default; monitor requires the verified human Monitor path. |
| caller_write_approvals | no | Additionally allow Write and Edit requests (including outside the repository) to be approved one exact action at a time. Requires caller_approvals and a version that qualifies it; defaults off, in whi | |
| group | string | no | Optional label (1-128 chars), inherited by any nested dispatch this task's own agent makes through harflow_mcp (unless that dispatch gives its own). Useful for tagging a family of dispatches you want |
| title | string | no | Optional short human-readable label (at most 90 characters after trimming, no control characters) shown for this task in the Monitor. Purely informational. It is never inherited from a calling task: a |
| workflow | string | no | — |
| auto_continue | no | Integer 0–10 (default 0). Opt in to that many fresh successors after backend-confirmed recoverable cutoffs and observed process/output settlement. Supported cutoff profiles are backend-owned; missing | |
| notify | string | no | Configured destination alias, direct callers only. Intent is durable before spawn. Receivers must deduplicate stable IDs; retries resume only while a server is active. |
| extra_writable_paths | array | no | Optional bounded list of existing absolute local file or directory paths. At most 16 paths, 4096 UTF-8 bytes each and 16384 bytes total. Relative paths, missing/special paths, filesystem roots and can |
| allowed_commands | array | no | Optional list of Bash command patterns granted for this run only, so a task can run specific commands (a test runner, a build) without unrestricted. At most 16 patterns, 256 bytes each and 2048 in t |
| resources | array | no | Optional list of 1-16 resource names (lowercase letters, digits and . _ : -, at most 64 characters, starting with a letter or digit; for example ios-simulator or repo:main) this task needs exclu |
List bounded permission request metadata for exactly one task or root workflow tree.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | no | — |
| workflow_run_id | string | no | — |
| cursor | string | no | — |
| limit | no | — |
Read the complete exact action for an authorized direct caller.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| request_id | string | yes | — |
Reserve approve/reject for one exact action, with stale/duplicate answer protection.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| request_id | string | yes | — |
| action_digest | string | yes | — |
| owner_revision | yes | — | |
| decision | string | yes | — |
| interaction_owner | string | no | — |
Preview standalone task permissions without launching an agent or allocating scratch.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| repo_path | string | yes | — |
| backend | string | no | — |
| freedom | string | no | — |
| network | no | — | |
| caller_approvals | no | — | |
| extra_writable_paths | array | no | — |
| caller_write_approvals | no | — | |
| allowed_commands | array | no | — |
Durably subscribe to one task's observed terminal settlement using a configured alias.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| notify | string | yes | — |
Read bounded subscription delivery states; this never establishes task success.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
Stop future attempts for a task/alias (or all aliases); in-flight requests may finish.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| notify | string | no | — |
Report a dispatched task's current state without blocking.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | Identifier returned by start_task or resume_task. |
| include_tail | boolean | no | If True, include the full last_output_tail (20 lines × 2000 chars each); if False (default), the tail is omitted for running/completed tasks and shortened to 5 lines × 500 chars for failed tasks. The |
| brief | no | Return an 8 KiB display projection with bounded summary/outcome, the last three activities and detail references. Overrides include_tail. Full reads remain available. | |
| include_successors | no | Fetch the first authorized page of successor IDs. Defaults false: ordinary polling includes a detail reference without scanning retained task history. |
Wait for a task to finish, giving up after a timeout without disturbing the run.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | Identifier of the task to await. |
| timeout_seconds | no | How long to hold, 0..3600 (default 55; 0 observes once without sleeping). Honoured as asked, never silently clamped; out of range is INVALID_PARAMS. Breaking change: values above 3600 are now rejected | |
| include_tail | boolean | no | If True, include the full last_output_tail (20 lines × 2000 chars each); if False (default), the tail is omitted for running/completed tasks and shortened to 5 lines × 500 chars for failed tasks. |
| until | string | no | "timeout" (default) returns when the task settles or timeout_seconds elapses. "terminal" holds until the task settles, bounded only by a 6 hour server ceiling; timeout_seconds must then be omitted (pa |
Observe 1..32 unique task IDs, returning compact statuses in requested order.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_ids | array | yes | — |
| mode | string | no | — |
| timeout_seconds | no | — | |
| until | string | no | — |
Start a fresh session from a conclusively settled standalone task.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| extra_prompt | string | no | — |
| notes_path | string | no | — |
| background_idle_seconds | no | — | |
| allowed_commands | array | no | — |
Continue a finished task's session with follow-up instructions.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | A task that has finished; its session is the one resumed. |
| followup_prompt | string | no | What the agent should do next, with the prior context intact. Write it in plain prose, as with start_task's prompt: preserve spaces and word boundaries and never concatenate words to shrink the payloa |
| max_turns | integer | no | Cap on agent turns, where the backend supports one. |
| background_idle_seconds | no | — | |
| network | no | Optional boolean overriding the parent run's network setting for this run only: None (the default) inherits what the parent was dispatched with — which is also the historical default for a record writ | |
| allowed_commands | array | no | None (the default) inherits the parent's saved command grants; a list replaces them for this run after the same validation and nested-subset check as start_task, which documents the grammar and limits |
List dispatched tasks, oldest first.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| status | string | no | Optional filter — running, completed, failed, timed_out or cancelled. |
| backend | string | no | Optional filter by harness, e.g. "codex". |
| group | string | no | — |
| title_contains | string | no | — |
| repo | string | no | — |
| since | string | no | — |
| limit | no | — | |
| fields | array | no | — |
| cursor | string | no | — |
| waiting_for_resource | string | no | Optional resource name: only launches that have not started and are still queued for it (launch_state: "waiting_for_resources"). Uses the unpaged listing. |
Stable newest-first persisted task headers; inspect a task for reconciled status/output.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| limit | integer | no | — |
| cursor | string | no | — |
| active_only | boolean | no | — |
| session_id | string | no | — |
| task_ids | array | no | — |
| predecessor_task_id | string | no | — |
Stop a running task, terminating the agent and any processes it spawned, and cascade the same stop to any live descendants this bridge can find.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | Identifier of the task to stop. |
Fetch normalized events from a task's event log.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | Identifier of the task whose events to read. |
| limit | integer | no | Maximum number of events to return, capped at 200. Default is 50. |
| before_seq | integer | no | Exclusive upper bound — return events with seq < before_seq (older events). Cannot be used with after_seq. |
| after_seq | integer | no | Exclusive lower bound — return events with seq > after_seq (newer events). Cannot be used with before_seq. |
| kinds | array | no | Optional filter by event kind. Must be a non-empty list of known kinds from task_started, assistant_text, assistant_delta, tool_call, tool_result, user_message, usage, notice, `task_fi |
| recovery_cursor | string | no | — |
Read one bounded recent or older activity page using an opaque stable cursor.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| limit | integer | no | — |
| cursor | string | no | — |
| kinds | array | no | — |
Add a message to a running live-input task, as if the user had typed it mid-run — or queue it for the next continuation of a task that has no live input.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | A running task whose live_input is true (claude started without max_turns, or any antigravity task); or any other standalone task, running or settled, whose message waits for its next continuation |
| text | string | yes | The message, stored and delivered exactly as submitted — keep plain prose intact, as with start_task's prompt: spaces and word boundaries preserved, never words concatenated to save tokens. Blank or a |
Withdraw a follow-up that send_message queued for a task's next continuation.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | The task the message was queued on. |
| message_id | string | yes | The message_id send_message returned (also listed in the task's queued_followups). |
List inner saved definitions; any incompatible or unavailable source refuses the whole list.
Takes no parameters.
Read a saved workflow definition and revision.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
Validate and save an explicit workflow; expected_revision protects concurrent edits.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
| definition | object | yes | — |
| expected_revision | integer | no | — |
| expected_workflow_id | string | no | — |
Delete a qualified definition after scanning qualified referrers; historical snapshots remain available.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
Generate a workflow, or refine the current canvas into an unsaved validated proposal.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
| prompt | string | yes | — |
| repo_path | string | no | — |
| agent | object | no | — |
| fallbacks | array | no | — |
| definition | object | no | — |
| source | object | no | — |
Queue a message in the same workflow-builder conversation without saving its draft.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| run_id | string | yes | — |
| prompt | string | yes | — |
Publish a preview from the verified active builder task; never write a saved workflow.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| definition | object | yes | — |
| expected_draft_revision | integer | yes | — |
Preview resources, worktree plan, native ownership, permissions and fallback reasons without launching.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
| repo_path | string | yes | — |
| overrides | object | no | — |
| network | no | — | |
| worktree | object | no | — |
Start a durable workflow run; returns workflow_run_id rather than task_id.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
| prompt | string | yes | — |
| repo_path | string | yes | — |
| overrides | object | no | — |
| freedom | string | no | — |
| network | no | — | |
| expected_preview_hash | string | no | — |
| goal | string | no | — |
| notify | string | no | — |
| notify_events | array | no | — |
| worktree | object | no | — |
Subscribe a root workflow to configured webhooks (terminal, attention, or both).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| notify | string | yes | — |
| events | array | no | — |
Stop future workflow delivery attempts for one configured alias, or every alias.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| notify | string | no | — |
Read root workflow subscriptions and at most 50 recent delivery states; direct callers only.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
Revise a root run goal at a quiescent paused or awaiting-input checkpoint.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| run_id | string | yes | — |
| goal | string | yes | — |
| expected_revision | yes | — |
List recorded workflow runs.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| offset | integer | no | — |
| limit | integer | no | — |
Stable newest-first run headers, with bounded ancestor/session-owner lookups.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| limit | integer | no | — |
| cursor | string | no | — |
| active_only | boolean | no | — |
| related_run_id | string | no | — |
| run_ids | array | no | — |
Read a workflow run's execution state and task associations.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
Wait for a workflow run to need attention or finish; read-only and resumable.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| timeout_seconds | no | — | |
| until | string | no | — |
Losslessly read executions, decisions, checklist, technical_plan, definition, owner_contracts, builder_draft, generated_definition, question, reason or wait_reason.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| view | string | yes | — |
| cursor | string | no | — |
| limit | integer | no | — |
Stop scheduling while active steps settle.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
Resume a caller-owned run with instructions and optional explicit attempt grants.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| instructions | string | no | — |
| additional_attempts | integer | no | — |
| decision_id | string | no | — |
| allow_optional_review_skip | no | — |
Read an immutable input explicitly assigned to your active worker activation.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| execution_id | string | yes | — |
| source_run_id | string | no | — |
| task_id | string | no | — |
| cursor | string | no | — |
| limit | integer | no | — |
Inspect any fully settled node execution in your workflow, including older executions.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| execution_id | string | yes | — |
| task_id | string | no | — |
| view | string | no | — |
| cursor | string | no | — |
| limit | integer | no | — |
| before_seq | integer | no | — |
| after_seq | integer | no | — |
Explicitly recover a failed settled run or a root with a confirmed-dead scheduler.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| reason | string | yes | — |
| additional_attempts | integer | no | — |
Stop root scheduling and cascade cancellation to associated tasks and children.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
Register an existing regular-file metadata reference for the verified calling task.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| path | string | yes | — |
| kind | string | yes | — |
| note | string | no | — |
Page at most 50 artifact references with current stat observations, not content verification.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | — |
| cursor | string | no | — |
| limit | integer | no | — |
Page artifact references registered by the attempts of a workflow run and its linked child runs.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| workflow_run_id | string | yes | — |
| cursor | string | no | — |
| limit | integer | no | — |
| execution_id | string | no | — |
| task_id | string | no | — |
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under Apache-2.0
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Harflow or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 94 / 100
Release 0.0.14 · scanned 14 Nov 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score94
Findings13 threat · 1 capability
Capabilityhigh
Coveragesource
Verificationsource
Surface digeste5e0ccff50ca4f407887aa834ee5d3a19116d14deadd1330ae2d07a4a094163d
Score vectorthreat 100 → client 94
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (64 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
tests/test_clients.py (85 KB) was not read: over the file-count budget.
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| hold | no tags | |
| list_backends | no tags | |
| start_task | no tags | |
| list_approval_requests | no tags | |
| get_approval_request | no tags | |
| respond_approval_request | no tags | |
| preview_task_permissions | untrusted-input | untrusted-input: keyword "fetch" |
| subscribe | no tags | |
| get_task_notifications | no tags | |
| unsubscribe | no tags | |
| get_task_status | untrusted-input | untrusted-input: keyword "fetch" |
| wait_for_task | no tags | |
| wait_for_tasks | no tags | |
| continue_task | untrusted-input | untrusted-input: keyword "fetch" |
| resume_task | untrusted-input | untrusted-input: keyword "fetch" |
| list_tasks | no tags | |
| list_task_page | no tags | |
| recent_limit_hits | no tags | |
| cancel_task | no tags | |
| get_task_events | untrusted-input | untrusted-input: keyword "fetch" |
| get_task_event_page | no tags | |
| send_message | external-sink | external-sink: keyword "send_message" |
| prepare_task_conversation | no tags | |
| send_conversation_message | no tags | |
| control_conversation_message | no tags | |
| commit_conversation_result | no tags | |
| cancel_queued_message | no tags | |
| send_workflow_message | no tags | |
| cancel_workflow_message | no tags | |
| list_workflows | no tags | |
| get_workflow | no tags | |
| save_workflow | no tags | |
| delete_workflow | no tags | |
| workflow_builder | no tags | |
| followup_workflow_builder | no tags | |
| apply_workflow_draft | no tags | |
| preview_workflow_run | no tags | |
| start_workflow | no tags | |
| subscribe_workflow | no tags | |
| unsubscribe_workflow | no tags | |
| get_workflow_notifications | no tags | |
| get_workflow_steering | no tags | |
| get_workflow_steering_history | no tags | |
| steer_workflow | no tags | |
| replan_workflow_steering | no tags | |
| revise_workflow_goal | no tags | |
| list_workflow_runs | no tags | |
| list_workflow_run_page | no tags | |
| workflow_limit_hits | no tags | |
| get_workflow_status | no tags | |
| wait_for_workflow | no tags | |
| get_workflow_run_detail | no tags | |
| pause_workflow | no tags | |
| resume_workflow | no tags | |
| read_workflow_assigned_input | no tags | |
| inspect_workflow_node | no tags | |
| recover_workflow | no tags | |
| cancel_workflow | no tags | |
| register_artifact | no tags | |
| get_task_artifacts | no tags | |
| get_workflow_artifacts | no tags | |
| acquire_slot | no tags | |
| release_slot | no tags | |
| slot_status | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
flow-1
Untrusted input can drive an external action even though no sensitive source is exposed.
Supply chain and provenance
Integrityfirst-seen
Surface digeste5e0ccff50ca4f40
Scanned2026-11-14
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 0.0.14. Pin a release below and the command asks for exactly that one.
0.0.14
14 Nov 2026
latestNo release note was published with this version.
0.0.13
10 Nov 2026
No release note was published with this version.
0.0.12
9 Nov 2026
No release note was published with this version.
0.0.11
6 Nov 2026
No release note was published with this version.
0.0.10
5 Nov 2026
No release note was published with this version.
0.0.9
4 Nov 2026
No release note was published with this version.
0.0.8
4 Nov 2026
No release note was published with this version.
0.0.7
2 Nov 2026
No release note was published with this version.
0.0.6
1 Nov 2026
No release note was published with this version.
0.0.5
29 Oct 2026
No release note was published with this version.
0.0.4
undated
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Harflow can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.